Gastric, Pancreatic, and Intestinal Function

Peter G. Hill, B.Sc., Ph.D., F.R.C.Path., Dip.Hlth.Mgt.*

The stomach, intestinal tract, and pancreas are closely related, both anatomically and functionally, and symptoms, such as diarrhea or malabsorption, may be associated with diseases or disorders of any of these organs. It is therefore appropriate to discuss them together. Advances in imaging techniques and improvements in endoscopic procedures have led to enormous changes in the investigation of gastrointestinal (GI) and pancreatic function so that many laboratory tests, once considered important, have now been superseded.

In this chapter, the anatomy and physiology of the GI tract, and the normal processes of digestion and absorption, are briefly reviewed. Disorders of the stomach, pancreas, and intestine associated with malabsorption or diarrhea, in which the laboratory can play a role in diagnosis and monitoring, are discussed. The chapter concludes with an overview of GI regulatory peptides and neuroendocrine tumors in which GI symptoms are prominent, and with two sections that present more integrated approaches to the problems of investigating malabsorption and diarrhea.

Introduction to Anatomy and Physiology of the Gastrointestinal Tract ¹⁷⁶

The major organs of the GI tract include the stomach, the small and large intestines, the pancreas, and the gallbladder, all of which are involved in the digestive processes that commence with the ingestion of food and water and culminate in the excretion of feces.

Anatomy

The GI tract is a tube, 8 meters in length, beginning with the mouth and ending with the anus. The esophagus, which is about 25 cm in length, is a muscular tube connecting the pharynx to the stomach. For this chapter, the key organs are the stomach, intestines, and pancreas.

The stomach consists of three major zones: the cardiac zone, the body, and the pyloric zone (Figure 51-1). The upper cardiac zone, which includes the fundus, contains mucus-secreting surface epithelial cells, which also secrete group II pepsinogens and several types of endocrine secreting cells. The body of the stomach contains cells or cell groups of many different types: (1) surface epithelial cells, which secrete mucus; (2) parietal (oxyntic) cells, which secrete hydrochloric acid and intrinsic factor; (3) the chief, zymogen, or peptic cells, which secrete group I and II pepsinogens; (4) enterochromaffin cells, which secrete serotonin; and (5) several types of endocrine secreting cells. The pyloric zone is subdivided into the antrum (which is approximately the distal third of the stomach), the pyloric canal, and the sphincter. The cells of the pyloric zone secrete mucus, group II pepsinogens, serotonin, gastrin, and several other hormones but no hydrochloric acid.

Figure 51-1 Schematic drawing of the stomach, with major zones.

Processes of Digestion And Absorption ⁵²

The total quantity of fluid absorbed each day by the gut is estimated to be about 9 L, which is composed of 2 L oral intake, 1.5 L saliva, 2.5 L gastric juice, 0.5 L bile, 1.5 L pancreatic juice, and 1 L intestinal secretions. More than 90% of this fluid is absorbed in the small intestine. The maximal absorptive capacity for fluid is probably at least 20 L. Several hundred grams of carbohydrates, about 100 g of fat, and 50 to 100 g of amino acids are absorbed daily in the small gut, but maximal absorptive capacity is believed to be at least 10 times greater. This considerable reserve capacity may compensate for mild to moderate degrees of dysfunction induced by disease processes, at least in the early phases. The efficiency of absorption is due to the unique features of the absorptive surface of the bowel and the relationship of the epithelial cells to the underlying rich vascular plexus and the lymphatic vessels.

Digestion of ingested food takes place both within the lumen of the small intestine and at the mucosal (brush border) surface. Defects of digestion may occur at one or more stages in the process. The terms maldigestion and malabsorption refer to different functional abnormalities. Maldigestion is a dysfunction of the digestive process that may occur at various sites in the GI tract. For example, hypoacidity in the stomach will reduce peptic digestion of protein; hyperacidity of the duodenum (e.g., due to overproduction of gastrin by tumor in the Zollinger-Ellison syndrome) can inactivate pancreatic enzymes; loss of brush border enzymes in the small intestine, because of any of a variety of processes, can prevent oligosaccharides and disaccharides from being further hydrolyzed; pancreatic insufficiency will reduce intraluminal enzyme activity in the small gut, causing maldigestion of fats and proteins. By contrast, malabsorption is strictly a dysfunction of the absorptive process in the small gut due to loss of absorptive epithelial cells caused, for example, by gluten, inflammation, infection, surgical resection, and infiltration. Various transport defects also lead to malabsorption of specific substances (e.g., glucose-galactose malabsorption or zinc deficiency in the congenital disorder acrodermatitis enteropathica). In clinical practice, however, the term malabsorption is often used to encompass all aspects of impaired digestion and absorption. As Figure 51-3 shows, absorption of the different nutrients proceeds at different rates and at different sites within the small bowel.

Figure 51-3 The location of small intestinal absorption. (Modified from Morris JA, Selivanov V, Sheldon GF. Nutritional management of patients with malabsorption syndrome. Clin Gastroenterol 1983;12;463-74, based on Borgstrom B, et al. Studies of intestinal digestion and absorption in the human. J Clin Invest 1957;36:1521-36.)

In the following three sections, the digestion and absorption of fats, carbohydrates, and proteins will be discussed separately. It must be remembered, however, that a complex interplay takes place among nutrients, regulatory peptides, enzymes, gallbladder and pancreatic function, and bowel motility, leading to an integrated absorptive process that commences with the ingestion of food and culminates in the excretion of feces.

Digestion and Absorption of Carbohydrates

After the action of salivary and pancreatic α-amylases on dietary starch and glycogen, the carbohydrate content of the small intestine consists of newly formed maltose; ingested monosaccharides; dietary disaccharides, such as lactose, sucrose, maltose, and trehalose; oligosaccharides, such as dextrins and maltotriose; and indigestible oligosaccharides and polysaccharides, such as cellulose, agar, and other oligosaccharide dietary fibers.

The brush border enzymes with disaccharidase and oligosaccharidase activity are listed in Table 51-1. The sucrase-isomaltase complex comprises most of the sucrase, isomaltase, and maltase (80%) activity of the small intestine. It hydrolyzes sucrose to its constituent monosaccharides, cleaves glucose from α-limit dextrins with 1,6 bonds, and hydrolyzes maltose. The activity of the complex is fourfold to fivefold greater in the jejunum than in the ileum. Changes in diet have a marked effect on the expression of the complex; starvation leads to a rapid decline in activity, which is rapidly restored on refeeding. All small intestinal saccharidases may decrease with infection or inflammation of the small bowel to the extent that carbohydrate malabsorption occurs, leading to diarrhea, flatulence, and weight loss. Paradoxically, diabetes mellitus causes a striking increase in sucrase-isomaltase activity; an increase is also observed in monosaccharide and amino acid transport. The lactase–phlorizin hydrolase complex is the only brush border enzyme able to hydrolyze lactose and therefore is essential for the survival of mammals early in life.

TABLE 51-1

Brush Border Oligosaccharidases

Enzyme or Complex	Principal Substrate(s)	Products
β-Glycosidase Complex
Lactase (EC 3.2.1.23)	Lactose	Glucose + galactose
Glycosylceramidase (EC 3.2.1.45-46) [also called phlorizin hydrolase (EC 3.2.1.62)]	Galactosyl and glucosyl-β-ceramides	Galactose/glucose + ceramides
Sucrase-Isomaltase Complex
Sucrase-(maltase) (EC 3.2.1.48)	Maltose/sucrose	Glucose/fructose + glucose
Isomaltase-(maltase) (EC 3.2.1.10)	1,6-α-linkages in isomaltose and α-dextrins Maltose	Glucose Glucose
Trehalase (EC 3.2.1.28)	Trehalose	Glucose
Glucoamylase Complex (EC 3.2.1.20)
Glucoamylase-(maltase)-1 and glucoamylase-(maltase)-2 (have similar substrate specificities)	1,4-α-linkages at nonreducing ends of amylose, amylopectin, glycogen, and straight chain 1,4-α-glucopyranosyl oligomers, including maltose	Glucose

From Semenza G, Auricchio S, Mantei N. Small-intestinal disaccharidases. In: Scriver CR, Beaudet AL, Sly WS, Valle D, eds. The metabolic and molecular bases of inherited disease, 8th edition. New York, NY: McGraw-Hill, 2001:1623-50.

This complex also has glycosylceramidase, β-glycosidase, and phlorizin hydrolase activities. Infectious and inflammatory diseases greatly reduce lactase–phlorizin hydrolase activity, leading to symptomatic intolerance to milk. Recovery of enzyme activity following intestinal disease may be slow. The activity of the complex is resistant to starvation. The developmental regulation of lactase is discussed later in the section on disaccharidase deficiencies. Also present in the brush border is the α-glucosidase maltase-glucoamylase, which removes individual glucose molecules from the nonreducing end of α(1,4) oligosaccharides and disaccharides. This enzyme accounts for about 20% of the total maltase activity of the small intestine. Trehalase is also found in the brush border of the small intestine and hydrolyzes trehalose, an α(1,1) disaccharide of glucose found in yeast and mushrooms. The developmental pattern of trehalase appears to follow that of sucrase-isomaltase.

In addition to their actions on disaccharides, the brush border enzymes further hydrolyze the products of amylase action, including maltose, maltotriose, and α-limit dextrins. The brush border enzymes appear to act in an integrated manner in that a flow of substrate occurs from glucoamylase and isomaltase to sucrase producing the monosaccharides glucose, galactose, and fructose. These monosaccharides are transported into the enterocyte by facilitative transport systems, such as the Na⁺-dependent glucose (and galactose) transporter (SGLT1) and GLUT5 (one of the GLUT family of monosaccharide transporters), which transports fructose across the apical membrane of the enterocyte. Subsequently, absorbed glucose and fructose are transported across the basolateral membrane and out of the enterocyte and into the portal system by the GLUT2 transporter.

It is increasingly being realized that the limiting factor in carbohydrate digestion and absorption may be diffusion from the intestinal lumen to the membrane surface where the enzymes are localized. Normally, little disaccharidase activity is seen in the luminal contents. For most oligosaccharides (with the exception of lactose), hydrolysis is rapid, and transport is the rate-limiting step in reducing the concentration of monosaccharides and the osmotic load in the gut. When the transport system is operating at its maximum rate but monosaccharide concentration is still high, inhibition of hydrolases by their monosaccharide products (i.e., product inhibition) slows hydrolytic activity, keeping monosaccharide concentrations relatively constant, thereby controlling osmotic load and water concentration in the gut. The importance of this control is evident from the consequences of intestinal disorders in which ingested disaccharide is not split and absorbed, leading to increased fluid secretion into the gut and increased intestinal motility. Enteric bacteria ferment the unabsorbed sugars, producing hydrogen, carbon dioxide, and organic acids causing abdominal discomfort such as bloating, distention and cramping. Absorption of fermentation products may lead to metabolic acidosis. In the large bowel, the presence of CO₂ and organic acids decreases pH and keeps the osmolality high, so that water reabsorption is decreased. The result is an acidic, liquid stool. Normally, however, accumulation of monosaccharide products does not occur, because the transport system is sufficiently fast to remove them. Mucosal lactase activity is the lowest of all the disaccharidases; for lactose, the rate-limiting step in absorption is thought to be hydrolysis. Lactase activity is not increased by feeding large amounts of lactose, as is the case for maltase and sucrase. Lactase, maltase, and sucrase all show circadian rhythms in their activities; minimum and maximum activities may vary by a factor of 2.

Carbohydrate digestion is not always complete in the small intestine. Indeed, it is likely that some starch and sucrose normally pass undigested and unabsorbed into the colon. It has been estimated that colonic bacteria require 70 g of carbohydrate/d. Much of this is derived from endogenous sources, such as from glycoproteins in GI secretions, with the remainder coming from unabsorbed dietary carbohydrate and dietary fiber. Up to 15% of the carbohydrate from white bread reaches the colon, and the effects of indigestible oligosaccharides upon reaching the large bowel are well known. As was pointed out earlier, bacterial action creates short chain fatty acids, which are rapidly absorbed by the colonic mucosa and are thought to provide fuel for the colonocyte. Starch and oligosaccharides are osmotically active and draw water into the gut. The colon, however, can absorb up to four times the normal colonic water load; for this reason, diarrhea is not always present in oligosaccharide malabsorption.

Digestion and Absorption of Lipids

The recommended daily dietary fat intake in Europe and North America is 70 to 95 g. Less than 5 g/24 h is recovered in the feces, indicating the overall efficiency of the normal processes of fat digestion and absorption. Most dietary fat is in the form of long chain triacylglycerols (triglycerides). Pancreatic lipase is quantitatively the most important hydrolytic enzyme, but the contribution of gastric lipase to overall hydrolysis should not be underestimated. Gastric lipase is secreted by the gastric mucosa and normally accounts for up to 17.5% of fatty acids released from triglycerides following a meal.¹⁸ The enzyme has a wide pH optimum and is active in both the stomach and the duodenum. This nonpancreatic lipase may have a significant role in lipid digestion when pancreatic function is impaired and in the neonatal period before pancreatic lipase activity is fully developed. A lingual lipase is also present, secreted by the tongue, but is thought not to be of much significance normally in humans. Fats first are emulsified in the stomach by its churning action and are stabilized by interaction with luminal lecithin and protein fragments. The lingual and gastric lipases do not require bile salts or cofactors to function; they have a pH optimum of 3 to 6, and their action produces 1,2-diacylglycerols and fatty acids. These products further stabilize the surface of the triglyceride emulsion and in the duodenum promote the binding of pancreatic colipase. In addition, the liberated fatty acids stimulate release of CCK from the duodenal mucosa.

Pancreatic lipase, in the presence of bile salts and colipase, acts at the oil-water interface of the triglyceride emulsion to produce fatty acids and 2-monoacylglycerols. Colipase is secreted in pancreatic juice as an inactive proenzyme, which is converted to the active form by trypsin. Other significant enzymes involved in the breakdown of fats within the intestinal lumen are cholesterol ester hydrolase, phospholipase A2, and a nonspecific bile salt–activated lipase.

Only a small proportion of ingested triacylglycerol is completely hydrolyzed to glycerol and fatty acids. These products form micelles with bile salts and lysophosphoglycerides; the micelles convey the nonpolar lipid molecules from the lumen to the epithelial cell surface and dissociate there to produce a high concentration of monoacylglycerols, lysophosphoglycerides, and fatty acids, which are absorbed into the mucosal cell. Absorption involves both passive and active transport processes and is facilitated by a fatty acid–binding protein in the cytosol of the cell that has a high affinity for fatty acids. Within the cell, triacylglycerols are resynthesized from the absorbed 2-monoacylglycerols and fatty acids. The triacylglycerols, together with phospholipids, cholesterol and its esters, fat-soluble vitamins, and a specific apolipoprotein, are formed into chylomicrons, which are then released by exocytosis into the lymphatic system of the small bowel. The absorption of long chain fatty acids is facilitated by transmembrane fatty acid transport proteins.

From the lymphatics, chylomicrons enter the bloodstream via the thoracic duct and are distributed to the liver, adipose tissue, and other organs. Medium and short chain fatty acids (chain length <12 carbon atoms) in mixed triglycerides are preferentially split by lipases and pass into the aqueous phase, from which they are rapidly absorbed. Medium chain triglycerides can be absorbed without complete lipolysis and in the absence of bile. They do not require micellar solubilization and are transported from the intestinal epithelial cells predominantly via the hepatic portal vein. Figure 51-4 summarizes the processes involved in fat absorption and conditions that compromise the efficiency of one or more stages in the process of fat digestion and absorption leading to fat malabsorption.²⁴

Figure 51-4 Summary of the processes involved in fat absorption and malabsorption. (From Clark ML, Silk DB. Gastrointestinal disease. In: Kumar P, Clark M, eds. Clinical medicine, 6th edition. Edinburgh: WB Saunders, 2005:265-345.)

Digestion and Absorption of Proteins

Average daily dietary intake of protein in North America is about 100 g compared with an estimated requirement for adults of 50 to 70 g. Another 50 to 60 g of protein enters the intestinal tract daily in GI secretions and from desquamated mucosal cells. Normal daily fecal loss of protein is about 10 g.

Protein digestion is initiated in the stomach by the action of pepsin in a highly acid medium. The acidity also denatures the protein, unfolding the polypeptide chains for better access by the gastric, pancreatic, and intestinal proteolytic enzymes. The polypeptides and amino acids produced in the stomach by the action of pepsin are potent secretagogues for hormones that stimulate the pancreas and intestine. Stimulated pancreatic secretion contains proenzyme forms of the proteolytic enzymes trypsin, chymotrypsin, elastase, exopeptidases, and carboxypeptidases. Proteolytic enzymes may be endopeptidases (e.g., pepsin, trypsin, chymotrypsin, elastase), which hydrolyze peptide bonds within the polypeptide chain, or exopeptidases, which hydrolyze peptide bonds of the terminal amino acids (enzymes such as carboxypeptidase and aminopeptidase). Stimulation of the intestine by GI hormones liberates several proteolytic enzymes from the brush border. One of them, enterokinase, selectively cleaves a hexapeptide from the N-terminus of trypsinogen to form trypsin. Trypsin then activates more trypsin (autocatalysis) and also converts other pancreatic proenzymes into their active forms. The action of the pancreatic enzymes on partially digested proteins within the lumen produces peptides that are 2 to 6 amino acid residues in length, as well as single amino acids. The peptides are largely hydrolyzed to single amino acids by the aminopeptidases and dipeptidases of the brush border before absorption, although some dipeptides and tripeptides are absorbed and are hydrolyzed to amino acids by cytosolic peptidases within the enterocytes. Multiple carrier systems with overlapping specificities for the 20 amino acids are involved in the transport of amino acids into the cells. Absorption of amino acids by these transport systems is faster in the jejunum than in the ileum. The amino acids pass across the enterocyte basolateral membrane by passive diffusion and by active transport systems, which are distinct from those at the brush border membrane. The underlying rich vascular plexus is drained by the portal circulation, and it is by this route that absorbed amino acids reach the liver and then the systemic circulation.

Individuals with achlorhydria or total gastrectomy have normal protein digestion and absorption because small intestinal function compensates for the lack of pepsin activity. Pancreatic and small intestinal diseases are the major causes of protein maldigestion and malabsorption. However, fecal loss of protein rarely becomes significant in pancreatic insufficiency until trypsin levels fall to about 10% of normal. Two rare disorders, trypsin deficiency and enterokinase deficiency, have far-reaching effects on the efficiency of protein digestion, as would be expected from their roles in the activation of proteolytic proenzymes. Mucosal diseases may affect protein assimilation through a number of mechanisms. Reduction in the number of mucosal cells decreases peptidase activity in the intestine and intestinal absorptive capacity for amino acids. Disease may increase the turnover of intestinal cells and the rate of desquamation. This cell loss, together with increased losses of plasma proteins from the damaged intestinal surface, can cause a negative nitrogen balance. Surgical resection of the intestine not only reduces the total intestinal absorptive surface but also may remove a segment of the gut that is specialized for absorption of certain nutrients (e.g., resection of the distal ileum removes the active transport system for the vitamin B₁₂–intrinsic factor complex). Resection may also alter intestinal motility, leading to stasis and bacterial overgrowth that can intensify a negative nitrogen balance. Also, rare hereditary defects in amino acid transporters (e.g., Hartnup’s disease) may produce distinct syndromes.

Stomach: Diseases and Laboratory Investigations

Growth in endoscopic procedures, with direct visualization of the interior of the stomach, has largely removed the need for the clinical laboratory to carry out analysis of gastric contents. Situations remain, however, in which the laboratory continues to play a significant role in diagnosing gastric diseases and in monitoring the effectiveness of treatment. This section describes peptic ulcer disease, tests for Helicobacter pylori (H. pylori), and measurement of basal acid output from the stomach.

Peptic Ulcer Disease and Helicobacter pylori64,123,158,180

Although the presence of spiral-shaped organisms in the stomach has been acknowledged for many years, it was only in 1985 that the association was described between H. pylori (known then as Campylobacter pylori) and peptic ulcer disease.¹²⁵ Most estimates suggest that the bacterium is present in the mucous layer of the stomach in half the population of the world. In Europe, 30 to 50% of adults, and in the United States, at least 20% of the adult population, are infected with the organism. In all cases, colonization with H. pylori causes a chronic inflammatory reaction in the gastric mucosa even when direct endoscopic observation of the mucosa appears normal. Carriers of the organism are at increased risk for gastric cancer (twofold to tenfold) and peptic ulcer (threefold to tenfold).¹² Some of this increased risk is due to infection with strains of the organism that produce the cytotoxic CagA protein. About 90% of gastric cancer patients are infected with H. pylori, compared with 40 to 60% of age-matched controls.^141,142 In a European study comparing the prevalence of H. pylori versus gastric cancer rates in 13 countries, a significant correlation was observed between infection rate and gastric cancer incidence and mortality.¹⁷⁵ It is, however, important to remember that although a large proportion of gastric cancer can be attributed to infection with H. pylori, only in a minority of infected subjects will the inflammatory reaction progress to gastric cancer. Gastric cancer rates in Western countries have declined in recent decades, but the incidence remains high in less developed countries.

At least 95% of patients with duodenal ulcer disease are infected with H. pylori, and eradication of the organism leads to healing of the ulcer and a reduction in relapse rates.¹⁴⁷ Eradication of H. pylori is now the recommended treatment for patients with duodenal or gastric ulcer who are H. pylori–positive. Effective combined antibiotic and acid suppression regimens (using PPIs) are available with eradication rates of up to 90% after first-line treatment.⁶⁴ However, increases in the prevalence of antibiotic resistance have led to the development of alternative treatment regimes to maintain high eradication rates.^122A

H. pylori infection predominantly affects the gastric mucosa, with the antrum usually the most densely colonized area. The reasons why a gastric mucosal infection predisposes to duodenal ulceration are complex and involve several pathways leading to increased acid production. Before there was an awareness of the role of H. pylori in the pathogenesis of peptic ulcer disease, vagotomy (surgical sectioning, or cutting, of the vagus nerve) was the mainstay of treatment as a means of reducing gastric acid output, thereby leading to an environment more conducive to healing of the ulcer.

Infection with the organism, with or without duodenal ulceration, leads to increases in both basal and meal-stimulated serum gastrin concentrations, principally due to an increase in gastrin-17.¹³⁴ Basal acid output is increased in H. pylori–positive subjects (Figure 51-5) and resolves completely after successful eradication of the organism. Hypergastrinemia is believed to be only one of the mechanisms leading to increased acid output. Studies using the neuropeptide gastrin-releasing peptide (GRP) suggest that impairment of inhibitory control mechanisms that regulate acid production may be responsible for the increased acid output associated with H. pylori infection.⁴⁹ In addition to stimulating G cells of the antrum to release gastrin, which leads to acid secretion by parietal cells, GRP activates neuroendocrine pathways that inhibit gastric acid secretion—an effect that is mediated via peptides (including cholecystokinin and secretin) that stimulate release of the inhibitory peptide somatostatin from the gastric mucosa.

Figure 51-5 Basal acid output in healthy volunteers and patients with duodenal ulcers (DUs) of varying Helicobacter pylori status (•, H. pylori positive). Compared with H. pylori–negative healthy volunteers, basal acid output is increased in H. pylori–positive healthy volunteers (P < 0.05), H. pylori–positive patients with DU (P < 0.0001), and patients with DU 1 month after eradication (P < 0.01). (From El-Omar EM, Penman ID, Ardill JES, Chittajallu RS, Howie C, McColl KE. H. pylori infection and abnormalities of acid secretion in patients with duodenal ulcer disease. Gastroenterology 1995;109:681-91, with permission from the American Gastroenterological Association.)

H. pylori produces urease, and hydrolysis of endogenous urea to bicarbonate and ammonia may create a more hospitable microenvironment for survival of the organism in the stomach. Mammalian cells do not hydrolyze urea, and it was only in 1984 that “gastric urease” was associated with the presence of H. pylori.¹¹⁰ The ability of the organism to rapidly hydrolyze urea is the basis of urea breath tests and of direct urease tests on gastric biopsy samples.

Diagnostic Tests for H. pylori

Numerous invasive and noninvasive diagnostic tests for H. pylori have been described (Box 51-1), and many have been reviewed.⁶⁸

Box 51-1 Diagnostic Tests for Helicobacter pylori

Invasive Tests—Using Gastric Mucosal Biopsy Samples

Histology: microscopy after Giemsa or silver staining

Histology: microscopy after immunohistochemical staining

Direct urease test: biopsy incubated in urea/indicator solution; visual end point

Culture: incubation on suitable media for 4 to 10 days

Polymerase chain reaction: amplification of specific DNA sequences

Noninvasive Tests—Using Breath, Blood, Saliva, or Feces

Breath tests: rise in breath ¹⁴CO₂ or ¹³CO₂ after ingestion of ¹⁴C- or ¹³C-labeled urea

Serum tests: laboratory measurement of specific immunoglobulin (Ig)G antibody

Whole blood tests: point-of-care tests for specific IgG antibody

Saliva tests: detection of specific IgG antibody

Fecal tests: detection of specific antigen

All tests in the “invasive” group necessitate oral gastroduodenoscopy with biopsy of the gastric mucosa; false-negative results may occur as the result of sampling errors, as colonization may be patchy. The antrum is the preferred biopsy site, but multiple biopsies from the anterior and posterior walls of the antrum and the body of the stomach are recommended for maximum diagnostic accuracy of this group of tests. False negatives may also occur when biopsy specimens are taken during treatment with PPIs or within 2 weeks of stopping PPI therapy. These drugs alter the intragastric distribution of H. pylori and suppress its activity.¹¹⁸ During PPI therapy, biopsies should be taken from the body and fundus to prevent false negatives. PPIs can also lead to falsely negative urea breath test results. If PPIs cannot be withheld for at least 2 weeks before a breath test, a negative result must be interpreted with caution. Histamine H₂-receptor antagonists should be stopped at least 24 hours before a breath test. Antacids do not affect the test results.

Tests for H. pylori are required for the diagnosis of infection and to ascertain, in some situations, whether eradication therapy has been successful. High sensitivity is required to ensure that positives are not missed; similarly, high specificity is essential to prevent inappropriate use of eradication therapy. The Maastricht III Consensus Guidelines ¹²³ recommend a “test and treat” strategy in adults with appropriate dyspeptic symptoms younger than 45 years using a breath test or a stool antigen test. The age limit may vary depending on local prevalence and the age distribution of gastric cancer (e.g., in the United Kingdom, testing and treatment are now an option in any patients with uncomplicated dyspepsia, although for those aged 55 years and older with unexplained and persistent recent-onset dyspepsia alone, referral should be made for urgent endoscopy).¹³⁹ Successful eradication of H. pylori should be confirmed with the urea breath test or by a monoclonal antibody-based stool antigen test if urea breath tests are not available. Other national guidelines confirm the urea breath test as the preferred procedure, both for initial diagnosis and for confirmation of eradication.^22A,86,139 Testing to confirm eradication should be done at least 4 weeks after completion of the course of treatment.

Urea breath tests are simple to perform, and both sensitivity and specificity are greater than 95%. Urea labeled with ¹⁴C or ¹³C is given orally as a drink or a capsule to be swallowed with water; urease from gastric H. pylori rapidly hydrolyzes the ingested urea to produce labeled bicarbonate, which is absorbed into the blood and exhaled as ¹⁴CO₂ or ¹³CO₂. The principal advantages of the ¹³C-urea breath test over the ¹⁴C-urea breath test are the simplicity of breath collection and the avoidance of regulations and environmental issues related to the use and disposal of radioisotopes. In the ¹⁴C-urea breath test, CO₂ in expired air is trapped in methanolic hyamine hydroxide as the patient exhales through a straw, which should be fitted with a one-way valve to ensure that the patient does not suck the trapping solution into the mouth. A color change of an indicator (thymol blue) in the solution shows that the required quantity of CO₂ has been trapped. Scintillant is then added and ¹⁴CO₂ measured. In the ¹³C-urea breath test, the patient blows through a straw into an empty 15 mL tube, which is then capped. ¹³CO₂/¹²CO₂ ratios are compared for basal and postdose samples using isotope ratio mass spectrometry or alternative infrared measurement methods.106,117

In the stool test, specific H. pylori antigens are detected in microtiter plates coated with polyclonal or monoclonal antibodies. Sensitivity and specificity are lower than for the ¹³C-urea breath test. Monoclonal antibody tests are recommended for posteradication testing if the urea breath test is not available.¹²³

Although still widely available, serologic tests are recommended only in specific situations (e.g., when PPI therapy cannot be withheld, when a patient with a bleeding ulcer is investigated).¹²³ The systemic antibody response is variable, leading to equivocal results in some subjects; in subjects older than 50 years, diagnostic accuracy is unsatisfactory. Serology cannot be used to confirm eradication because of the slow decline in antibody levels after treatment. Laboratory-based enzyme-linked immunosorbent assays (ELISAs) and point-of-care tests are available to measure specific immunoglobulin (Ig)G antibodies in serum or whole blood samples. In younger subjects, laboratory-based tests generally perform well, although some have sensitivity and/or specificity less than 95%. Office-based serology testing has inadequate sensitivity and specificity and currently is not recommended.¹²³ Calculations based on reported diagnostic accuracy data show that when these tests are used for diagnosis, as many as 28% of those receiving eradication therapy are being treated as a result of false-positive test results.²³

Determination of Basal Acid Output 35,91

Documentation of increased basal acid output (BAO) in gastric juice provides strong evidence that a high serum gastrin concentration is caused by Zollinger-Ellison syndrome. Therefore, this test is used in patients with duodenal ulceration and a raised serum gastrin concentration. The test is not appropriate in patients with atrophic gastritis. Pernicious anemia, which also causes hypergastrinemia, should be excluded before BAO is assessed. PPIs must be stopped for at least 14 days, and H₂-receptor antagonists for at least 3 days, before the test. H. pylori as a cause of increased serum gastrin should be excluded before BAO is estimated.

A basal condition, in the context of gastric analysis, is one in which the patient is at complete rest and is not exposed to any visual, auditory, or olfactory stimuli. Such a condition is maintained during sleep. It is for this reason that a 12 hour overnight collection of gastric juice has been used traditionally for the determination of BAO. Such an approach has the disadvantage that the patient is exposed to significant discomfort because of the need to retain the tube overnight, while sitting upright and slightly turned to the left to prevent loss of gastric contents into the duodenum. Close supervision throughout the entire night is necessary. A satisfactory alternative is the collection of gastric juice for 60 minutes after the patient has had a restful night’s sleep in a quiet separate room. After waking, the patient must remain fasting; smoking and exercise must be avoided before and during the test.

Collection of Gastric Juice

A gastric tube is inserted orally, or nasal intubation may be used if the patient has a hyperactive gag reflex. X-ray or fluoroscopic confirmation that the tip of the radiopaque tube is in the lowest portion of the stomach is necessary. Ten or 15 minutes after the patient has become calm and adjusted to the presence of the tube, the patient is positioned with the trunk upright and inclined slightly to the left. Gastric juice is then aspirated and discarded. After checking that no further juice can be aspirated, note the time, and collect and transfer to plastic bottles all gastric juice that can be aspirated over the next 60 minutes. The patient must be asked to expectorate all saliva during the collection period. The total volume of collected juice is recorded and free acid determined by titration as described in the following section.

Determination of Free Hydrochloric Acid in Gastric Juice

1. Measure and transfer a convenient volume of gastric juice (e.g., 5.0 to 10.0 mL) into a clean titration vessel. If the gastric juice contains food particles or mucus, centrifuge the sample or filter it through gauze.

2. Determine the pH of the gastric specimen with a pH meter. If the pH is above 3.5, no free acid is present. Such a specimen need not be titrated.

3. Titrate the sample with NaOH, 0.10 mol/L, to a pH of 3.5, using a pH meter.

Calculation:

If 5 mL of gastric specimen is titrated, the calculation becomes as follows:

Comments and Sources of Error

Titration to a pH of 3 to 3.5 detects essentially all free hydrochloric acid. (Because HCl is the only strongly ionized acid in gastric contents, this test is essentially a test for free HCl.) Titration beyond a pH of 3.5, as recommended by some, will overestimate the HCl concentration to varying degrees, depending on the composition of the gastric residue. On the other hand, titration to pH 3.5 may underestimate the amount of free H⁺ secreted if some of these H⁺ ions are bound to or have reacted with other constituents of gastric contents. Thus no fully satisfactory procedure is available to measure accurately the true total amount of free acid secreted by the gastric mucosa.

The effectiveness of gastric aspiration may be compromised by both the position of the patient and the position of the tube in the stomach (likened to the position of a straw above or below the fluid level in a glass), by the loss of gastric fluid into the pylorus, and by regurgitation of pyloric contents into the stomach. Some evidence suggests that the exact position of the tube in the stomach does not appear to alter the recovery of gastric juice, because the stomach is not rigid and its walls contract, thus shifting the stomach contents toward the tip of the tube.

Interpretation

Gastric juice pH: if the pH is greater than 2.5, then it is unlikely that the raised gastrin is caused by Zollinger-Ellison syndrome.

BAO: Normal: male 0 to 10.5 mmol/h; female: 0 to 5.6 mmol/h

Zollinger-Ellison syndrome: 15 to 100 mmol/h, or greater than 5 mmol/h if acid-reducing surgery was performed previously.

A level of free acid greater than 15 mmol/h should prompt suspicion of gastrinoma but is not diagnostic; a level greater than 25 mmol/h with high serum gastrin is virtually diagnostic of Zollinger-Ellison syndrome.

Further discussion of the Zollinger-Ellison syndrome can be found in the section of this chapter on neuroendocrine tumors.

Intestinal Disorders and Their Laboratory Investigation

This section includes discussions of celiac disease, disaccharidase deficiency, bacterial overgrowth, bile salt malabsorption, inflammatory bowel disease, and protein-losing enteropathy, and the main laboratory investigations associated with diagnosing or monitoring these disorders.

Celiac Disease (Celiac Sprue, Gluten-Sensitive Enteropathy)53,54,124,162,181A

Celiac disease is sometimes called nontropical sprue, celiac sprue, or gluten-sensitive enteropathy.

Pathophysiology of Celiac Disease

Celiac disease occurs in genetically predisposed subjects as a consequence of an inappropriate T cell–mediated immune response to ingestion of gluten from wheat and to similar proteins in barley and rye.

The role of genetic factors in celiac disease has been recognized for many years; a 70% concordance for celiac disease has been reported in identical twins, and typically 10% of first-degree relatives of an affected individual will be found to have the disease. Only recently has the major genetic component been localized to the human leukocyte antigen (HLA) region of chromosome 6. Approximately 95% of subjects with celiac disease express a specific HLA heterodimer (HLA DQ2 α/β heterodimer). Most Caucasian populations have a high frequency (20 to 30%) of DQ2, but only a small minority will develop celiac disease.

The external trigger to the development of celiac disease in genetically susceptible individuals is found in gluten, which is the complex group of proteins present in wheat that form a sticky mass when dough is washed with water and the starch is removed. All proteins (and peptides) that are toxic to the small bowel mucosa in subjects with celiac disease contain large amounts of glutamine. The major toxic proteins of wheat are the gliadins, with homologous proteins (the hordeins and secalins) occurring in barley and rye, respectively. The gliadins are a large family of proteins accounting for about 50% of the wheat protein.

Recent evidence indicates that oats do not lead to an immune-mediated response nor to mucosal damage in subjects with celiac disease.⁹⁰ Addition of oats to the list of permitted cereals increases choices and would be welcomed by most subjects with celiac disease. However, if oats are to be introduced into the diet, they must be obtained from a reliable source to ensure no contamination from wheat, barley, or rye proteins at any stage in the process from harvesting to packaging.

The identification in 1997 of small bowel tissue transglutaminase (tTG)-2 as the autoantigen of celiac disease ³⁶ was a major step forward in understanding the pathogenesis of this disorder. The tissue transglutaminases are a family of calcium-dependent cytoplasmic enzymes that are released from cells during wounding. Although their physiologic role is incompletely understood, they are able to catalyze the cross-linking of proteins, leading to stabilization of the wound area. Expression of the enzyme is increased during apoptosis and in active celiac disease. It selectively cross-links or deamidates protein-bound glutamine residues. Deamidation of gliadin peptides by the enzyme enhances their binding to HLA DQ2/DQ8 and increases recognition of these peptides by gut-derived T cells from subjects with celiac disease.¹⁶³ The pathogenesis of the disease is therefore believed to involve an interaction between tissue transglutaminase and gliadin peptides in genetically susceptible individuals.

The toxic cereal proteins lead to intestinal epithelial damage, releasing tissue transglutaminase. Cross-linking by the enzyme produces gliadin-gliadin or gliadin-enzyme complexes, unmasking new antigenic epitopes that bind to HLA DQ2 molecules on the antigen-presenting cells, producing an immune response by gut-derived T cells. The characteristic enteropathy is then induced by the release of interferon-γ and other proinflammatory cytokines, as outlined in Figure 51-6.

Figure 51-6 Hypothetical scheme for interaction between intestinal protein processing and the specific immune system in celiac disease. Dietary gluten is exposed to intraluminal digestive enzymes (including prolyl endopeptidases), producing stable 33-mer peptide-containing overlapping proline-rich epitopes with the PQPQLPY central motif. 33-mer peptide is absorbed intact by transcellular or paracellular routes into the lamina propria, and exposure to tTG results in deamidation of glutamine residues to glutamic acid. Modified peptide is presented to specific CD4⁺ T cells in association with HLA DQ2 or DQ8. Activated IFN-γ–producing T-effector cells return to the mucosa and produce characteristic villous atrophy and crypt hyperplasia. E, Glutamate; F, phenylalanine; IFN, interferon; L, leucine; P, proline; Q, glutamine; tTG, tissue transglutaminase; and Y, tyrosine. (From Mowat AM. Coeliac disease—a meeting point for genetics, immunology, and protein chemistry. Lancet 2003;361:1290-2.)

A 33-mer peptide of gluten appears to be the primary initiator of the inflammatory response.¹⁵⁶ It is resistant to breakdown by all gastric, pancreatic, and intestinal brush border membrane proteases, thus allowing it to reach the small intestine intact. After deamidation by tissue transglutaminase, it is a potent inducer of gut-derived human T-cell lines from patients with celiac disease. Homologs of the peptide are found in food grains that are toxic to patients with celiac disease, but are absent from nontoxic food grains. The peptide could be detoxified by exposure to a bacterial prolyl endopeptidase, suggesting a therapeutic strategy for celiac disease.¹⁵⁶

Increased intestinal permeability in untreated celiac disease that is reversible on withdrawal of gluten from the diet has been recognized since the early 1980s.⁷⁴ Evidence suggests that this may be mediated by increased expression of zonulin,⁵⁶ a protein that opens small intestinal tight junctions, or by decreased expression of intercellular epithelial cell adhesion molecules, such as Z0-1, catenin, and cadherin.¹⁴⁴ The zonulin pathway is now thought to play a significant role in the entry of allergens into the cell and hence in the autoimmune response.^54A

Clinical Considerations

Celiac disease is a common chronic disorder in Caucasian populations, with a prevalence of about 1%.122,186 It also occurs in northern Indian and North African populations. It is rare among Chinese, Japanese, and African-Caribbean people. It was previously considered to be a rare disorder in North America, but recent serologic and histologic evidence shows that the disease has been underdiagnosed, and that its prevalence is comparable, as might be expected, with that found in Europe.^55,77

A wide spectrum has been noted in the clinical presentation of celiac disease, with most diagnoses made in adult life. Classical celiac disease, presenting in infancy up to the age of 2 years, with failure to thrive, abdominal distention, and diarrhea, is now an uncommon presentation. The spectrum of presenting symptoms in adults has changed over the past 20 years, and frank malabsorption is now uncommon.87,185 Most adults present now with nonspecific symptoms; mild iron deficiency is common. A strong association with other autoimmune disease, especially with type 1 diabetes mellitus and autoimmune thyroid disease, has been reported. In type 1 diabetes, the prevalence of celiac disease is about 5%, and serologic screening to detect these cases has been advocated.⁸⁴ The initial presentation may be seen by a wide range of clinical specialties, as shown in Table 51-2. To make the diagnosis, there must be a high index of suspicion, along with awareness of the wide range of nonspecific symptoms and easy availability of serologic tests to select those patients in whom endoscopy is indicated to confirm the diagnosis.

TABLE 51-2

An Indication of the Wide Range of Clinical Specialties to Which a Child or Adult With Celiac Disease May Present

Clinical Specialty	Symptoms/Manifestations
General medicine	Tired all the time
Gastroenterology	Diarrhea, flatulence, weight loss
Hematology	Anemia
Obstetrics/gynecology	Infertility
Orthopedics	Fracture, osteopenia
Dermatology	Dermatitis herpetiformis, hyperkeratosis
Neurology	Peripheral neuropathy
Rheumatology	Arthropathy
Endocrinology	Short stature, thyroid disease
Diabetes	Diarrhea, anemia

Tests for Celiac Disease

Serologic tests have played a significant role in raising awareness of the high prevalence of this disorder, and appropriately standardized tests have high clinical sensitivity and specificity for diagnosis and for monitoring compliance with a gluten-free diet during treatment after diagnosis.

Table 51-3 compares the sensitivity and specificity of the four IgA class antibodies commonly used. Both antireticulin (ARA) and endomysial (EMA) antibodies are measured by indirect immunofluorescence, ARA on rat kidney sections and EMA on monkey esophagus or human umbilical cord sections. The presence of ARA indicates the need to measure antibodies with higher specificity before small bowel histology is recommended for confirmation of the diagnosis. ARA and EMA are tissue type–dependent methods that detect autoantibodies to tissue transglutaminase-2.¹⁰⁷

TABLE 51-3

Comparison of Serologic Tests for Celiac Disease

ELISA, Enzyme-linked immunosorbent assay; Ig, immunoglobulin.

Lack of standardization of assays for IgA-antigliadin antibodies (AGA) contributes to the variable diagnostic accuracy of this marker,⁸¹ but the sensitivity and specificity of AGA are poor, and use of this test should be abandoned. The sensitivity and specificity of current deamidated gliadin peptide antibody tests offer no advantages over tissue transglutaminase antibody (TGA) and may give significantly more false positives in subjects with liver disease.^80,136 The sensitivity of EMA in some reports is compromised by selection bias,¹¹⁵ but most larger series in which patients have not been selected for a biopsy on the basis of positive serology indicate that the true sensitivity for EMA is between 90% and 95% (i.e., 5 to 10% of subjects with celiac disease have a negative EMA at diagnosis). When carried out correctly, the assay has very high specificity (>99%); laboratories should monitor their performance, as small reductions in specificity will lead to a significant increase in the numbers of patients subjected unnecessarily to a small bowel biopsy. Table 51-4 shows the effects of test specificity on the numbers of true and false positives per 1000 subjects tested, and on the positive predictive value at disease prevalence (in the population tested) of 3%, assuming a sensitivity of 95%.

TABLE 51-4

Comparison of True and False Positives and Positive Predictive Values at Different Specificities, Assuming a Prevalence of Celiac Disease of 3% in the Population (n = 1000) Tested (Test Sensitivity of 95%)

Many commercial kits are now available to measure IgA antibodies against tissue transglutaminase [“transglutaminase antibody” (TGA)] using human recombinant tissue transglutaminase or purified human enzyme as antigen. Their role in the investigation of celiac disease has been reviewed.⁸⁰ Lack of standardization and differences in recombinant technology (e.g., the use of eukaryotic or prokaryotic organisms to produce tTG) can lead to variable performance.^11,126 Specificity should be evaluated using a large series of samples representative of those routinely tested, and procedures should be selected on the basis of high specificity (minimum 99%) and high sensitivity (>90%). The use of TGA has advantages over EMA. Unlike EMA, which may be subject to observer bias, TGA is a quantitative procedure that does not require the use of primate tissue. It can be automated and therefore is appropriate for larger numbers of samples. This test has replaced EMA as the antibody of choice for performing serologic testing and for assessing dietary compliance of subjects on a gluten-free diet.^14,79 As with all tests, not all kits perform to the same high standards, and laboratories should ensure that they participate in external quality assessment programs and select a well-validated method.^45,80 In view of the growing public interest in celiac disorder, further standardization of these kits is urgently needed.

For diagnosis, current guidelines require a jejunal biopsy with the characteristic villous atrophy, increased intraepithelial lymphocytes, and hyperplasia of the crypts.¹²⁴ Wider use of serology has led to the recognition of more cases with minimal histologic changes and to use of the terms potential and latent celiac disease.⁸³ Patients with positive EMA or TGA are described as having potential celiac disease when the jejunal biopsy has normal villi but shows subtle changes, such as increased numbers of intraepithelial lymphocytes. It is thought that in time these patients will develop a flat mucosa. The term latent celiac disease is used to describe patients who have at some time had positive antibodies and a flat mucosa that recovers on a gluten-free diet, but have also been found to have a normal mucosa while on a normal diet.

Subjects with selective IgA deficiency (IgA <0.05 g/L, incidence about 1 : 600) are at greater risk of celiac disease. It is therefore important to identify these individuals rather than risk excluding the diagnosis on the basis of a negative test for IgA class antibodies.²² When EMA is used, total serum IgA should be measured on all samples submitted for celiac serology to identify those patients with IgA deficiency. A TGA method should be used that can distinguish those with a “normal” concentration from the small proportion of subjects with very low concentrations, in whom total IgA measurement is indicated to confirm or exclude IgA deficiency.⁵⁷ When IgA deficiency is identified, IgG antibodies (IgG-EMA, IgG-TGA, and IgG anti-deamidated gliadin peptide) should be used as serologic screening tests.²¹ In view of the increased risk of the disorder in IgA deficiency, a small bowel biopsy should be considered in all IgA-deficient subjects with symptoms of celiac disease.²²

Strict adherence to a gluten-free diet leads to mucosal recovery in celiac disease and reduces the risk of bowel malignancy. TGA can be used as a marker for monitoring dietary compliance, in addition to its diagnostic role.¹⁴ Failure of symptoms to respond to a prescribed gluten-free diet may indicate (1) nonadherence to the diet, (2) other coexisting conditions (such as small bowel bacterial overgrowth, lactose intolerance, and microscopic colitis), or (3) the presence of refractory celiac disease. Refractory celiac disease is characterized by persistent villous atrophy with an increase in intraepithelial lymphocytes in the small bowel while the patient is on a long-term gluten-free diet. In both responsive and refractory celiac disease, celiac antibodies typically are decreased with dietary therapy and remain within reference intervals unless individuals are re-exposed to gluten.

Two types of refractory celiac disease may occur and are differentiated by the types of T-cell populations found in the intestinal mucosa—polyclonal in type I disease and clonal in type II disease. Differentiation of the two types is accomplished by immunohistochemical, flow cytometric, or T-cell receptor γ gene rearrangement analysis of intestinal mucosal T cells.^129A Type II celiac disease has an unfavorable prognosis and typically is viewed as a precursor to enteropathy-associated T-cell lymphoma (EATL).

With the availability now of serologic tests with high diagnostic accuracy, older tests used to investigate celiac disease should be abandoned. For example, there is no place now in routine use for the xylose absorption test. Dual sugar tests (e.g., cellobiose and mannitol) to assess small bowel permeability have a role in research studies but as yet have not established a place in the routine diagnosis or monitoring of celiac disease. Tests of fat malabsorption are also inappropriate in the diagnosis of celiac disease, although appropriate tests of pancreatic function (e.g., fecal elastase) may be indicated in patients diagnosed with celiac disease who fail to respond to a gluten-free diet.

Disaccharidase Deficiencies 155,171

The presence of the brush border disaccharidases is essential for carbohydrate absorption, and a reduction in their activity leads to carbohydrate malabsorption and intolerance. Carbohydrate malabsorption does not always lead to clinical symptoms, but when symptoms do occur (e.g., abdominal pain, flatulence, diarrhea) as a consequence of malabsorption, the patient is described as having carbohydrate intolerance.

Adult-type hypolactasia is the single most common absorptive defect in adults, with an incidence of 5 to 90% depending on the racial group, as shown in Table 51-5.

TABLE 51-5

Prevalence of Hypolactasia in Adults

Racial Group	Prevalence, %
Chinese	>90
American blacks	54-81
Asians	60-90
Greeks	60-78
Northern Europeans	5-30

Congenital Lactase Deficiency

Intestinal lactase is essential in infancy, and congenital lactase deficiency is a very rare disorder in which lactase activities in the mucosa are low or undetectable at birth. Severe diarrhea occurs as soon as milk is taken; stools have a low pH and contain large amounts of lactose and glucose produced by bacterial action on undigested lactose. A definitive diagnosis must be deferred until after maturation of the lactase synthesis system has occurred. In the interim, relief is dependent on adjustments to dietary composition that appear to reduce the severity of symptoms. An abnormal oral lactose tolerance test obtained a few months after birth could also be caused by congenital glucose-galactose intolerance (see discussion later in this chapter); the differential diagnosis requires performance of an oral glucose tolerance test in conjunction with the lactose tolerance test.

Adult-type Hypolactasia (“Acquired Lactase Deficiency”)

In most humans (and in all other mammals), expression of the enzyme decreases during childhood, and by adulthood lactase levels are 10% or less of those seen in infancy. If symptoms of flatulence, abdominal discomfort, bloating, or diarrhea occur after consumption of one or two glasses of milk or of a large portion of ice cream or yogurt, lactose intolerance should be suspected. Suspicion would be increased in a subject from an ethnic group with a high incidence of lactose intolerance (see Table 51-5).

Lactose intolerance may also occur as a result of reduced enzyme activity following diffuse intestinal damage from infection (giardiasis, bacterial overgrowth, or viral gastroenteritis), ulcerative colitis, celiac disease, and tropical sprue. This deficiency is usually reversible following recovery from the disorder.

Diagnostic Tests for Lactase Deficiency

Many methods have been proposed for detecting lactase deficiency (Box 51-2). Disaccharidase activities can be measured in homogenates of an intestinal biopsy.³² These assays are rarely required for routine diagnostic purposes, but when necessary (e.g., in investigations in infancy), they must be carried out by laboratories with expertise in these tests. Breath hydrogen is now the preferred test for diagnosing lactase deficiency. The use of hydrogen breath tests in disorders of carbohydrate absorption and in bacterial overgrowth has been reviewed.¹⁵⁹

Box 51-2 Methods for Detecting Lactase Deficiency

Lactase in mucosal biopsy

Oral lactose tolerance

Measure increase in plasma glucose

Measure increase in plasma galactose

Measure increase in breath H₂

Measure increase in breath ¹³CO₂

Oral Lactose Tolerance Tests

Oral tolerance tests, measuring the increase in plasma glucose or galactose following ingestion of lactose, have been used to diagnose lactase deficiency. The usual dose of lactose is 50 g in 200 mL water; lower doses should be used in children (2 g/kg, up to a maximum of 50 g). Multiple blood samples are collected over a 2 hour period and the peak increment in glucose (or galactose) noted. To exclude lactase deficiency, the increase above baseline for capillary plasma glucose should be greater than 1.1 mmol/L ¹³⁸ (20 mg/dL) or greater than 1.4 mmol/L (25 mg/dL) when venous plasma is used.²⁰ In a survey of laboratory practice in the United Kingdom, widely varying cutoffs were found to be in use (1.0 to 2.7 mmol/L) even with the same lactose dose.⁴⁴ The requirement for multiple blood samples and lack of procedural standardization suggest that these tests should be abandoned in favor of noninvasive breath hydrogen testing.

Hydrogen Breath Tests

Hydrogen is not an end product of mammalian metabolism, and breath hydrogen is derived from bacterial metabolism in the intestinal tract.¹¹⁴ Following lactose ingestion, the disaccharide normally will be split into its constituent monosaccharides and absorbed. With lactase deficiency, unabsorbed disaccharide will pass into the large bowel, and bacterial metabolism will produce hydrogen that is absorbed into the systemic circulation and exhaled in the breath. Breath hydrogen can be measured in end-expiratory breath with the use of laboratory or hand-held direct-reading electrochemical hydrogen monitors.

Patient Preparation: Appropriate patient preparation is essential to ensure stable baseline breath hydrogen levels (Box 51-3). Avoidance of wheat-based foods and fiber for 12 hours before the test minimizes the availability of substrates for bacterial metabolism in the large bowel. Fasting breath hydrogen is typically less than 5 ppm (5 µL/L), and concentrations greater than 20 ppm (20 µL/L) may be an indication of malabsorption or bacterial overgrowth.¹⁴³ Oral hygiene before ingestion of the substrate in hydrogen breath tests minimizes the production of hydrogen by oral bacteria. Brushing of teeth or use of an antibacterial mouthwash (e.g., 1% chlorhexidine) is recommended.¹²⁸ Mouthwash containing alcohol should not be used, because this may interfere with the measurement of hydrogen. Cigarette smoke contains high hydrogen levels; smoking therefore is not permitted immediately before or during the test.

Box 51-3 Protocol for Lactose Tolerance Test With Measurement of Breath Hydrogen