The clinical features depend on the extent and type of the malabsorbed nutrient. The common presenting features, especially in toddlers with malabsorption, are diarrhea, abdominal distention, and failure to gain weight, with a fall in growth chart percentiles. Physical findings include muscle wasting and the disappearance of the subcutaneous fat, with subsequent loose skinfolds (Fig. 330-1). The nutritional consequences of malabsorption are more dramatic in toddlers because of the limited energy reserves and higher proportion of calorie intake being used for weight gain and linear growth. In older children, malnutrition can result in growth retardation, as is commonly seen in children with late diagnosis of celiac disease. If malabsorption is untreated, linear growth slows, and with prolonged malnutrition, death can follow (Chapter 43). This extreme outcome is usually restricted to children living in the developing world, where resources to provide enteral and parenteral nutrition support may be limited. Specific findings on examination can guide toward a specific disorder; edema is usually associated with protein-losing enteropathy, digital clubbing with cystic fibrosis and celiac disease, perianal excoriation and gaseous abdominal distention with carbohydrate malabsorption, perianal and circumoral rash with acrodermatitis enteropathica, abnormal hair with Menkes syndrome, and the typical facial features diagnostic of the Johanson-Blizzard syndrome.

Figure 330-1 An 18 mo old boy with active celiac disease. Note the loose skinfolds, marked proximal muscle wasting, and distended abdomen. The child looks ill.

Many children with malabsorption disorders have very good appetites as they try to compensate for the fecal protein and energy losses. In exocrine pancreatic insufficiency, fecal losses of up to 40% of ingested protein and energy do not lead to malnutrition, as long as they are compensated by an increased appetite. In conditions associated with villous atrophy or inflammation (celiac disease, postinfectious enteropathy), fecal protein and energy losses are usually modest, but associated anorexia and reduced food intake results in malnutrition.

The nutritional assessment is an important part of clinical evaluation in children with malabsorptive disorders (Chapter 41). Long-term calcium and vitamin D malabsorption can lead to reduced bone mineral density and metabolic bone disease, with increased risk of bone fractures. Vitamin K malabsorption, irrespective of the underlying mechanism (fat malabsorption, mucosal atrophy), can result in coagulopathy. Severe protein-losing enteropathy is often associated with malabsorption syndromes (celiac disease, intestinal lymphangiectasia) and causes hypoalbuminemia and edema. Other nutrient deficiencies include iron malabsorption causing microcytic anemia and low reticulocyte count, low serum folate levels in conditions associated with mucosal atrophy, and low serum vitamin A and vitamin E concentrations in fat malabsorption.

Clinical history alone might not be sufficient to make a specific diagnosis, but it can direct the pediatrician toward a more structured and rational investigative approach. Diarrhea is the main clinical expression of malabsorption. Onset of diarrhea in early infancy suggests a congenital defect (Table 330-3). In secretory diarrhea due to disorders such as congenital chloride diarrhea and microvillus inclusion disease, the stool is watery and voluminous and can be mistaken for urine. Onset of symptoms after introduction of a particular food into a child’s diet can provide diagnostic clues, such as with sucrose in sucrase-isomaltase deficiency. The nature of the diarrhea may be helpful: explosive watery diarrhea suggests carbohydrate malabsorption; loose, bulky stools are associated with celiac disease; and pasty and yellowish offensive stools suggest an exocrine pancreatic insufficiency. Stool color is usually not helpful; green stool with undigested “peas and carrots” can suggest rapid intestinal transit in toddler’s diarrhea, which is a self-limiting condition unassociated with failure to thrive.

Table 330-3 DIARRHEAL DISEASES APPEARING IN THE NEONATAL PERIOD

CONDITION	CLINICAL FEATURES
Microvillus inclusion disease	Secretory watery diarrhea
Tufting enteropathy	Secretory watery diarrhea
Congenital glucose-galactose malabsorption	Acidic diarrhea
Congenital lactase deficiency	Acidic diarrhea
Congenital chloride diarrhea	Hydramnion, secretory watery diarrhea Metabolic alkalosis
Congenital defective jejunal Na^⁺/H^⁺ exchange	Hydramnion, secretory watery diarrhea
Congenital bile acid malabsorption	Steatorrhea
Congenital enterokinase deficiency	Failure to thrive, edema
Congenital trypsinogen deficiency	Failure to thrive, edema
Congenital lipase and/or co-lipase deficiency	Failure to thrive, oily stool
Enteric anendocrinosis (NEUROG 3 mutation)	Hyperchloremic acidosis, failure to thrive

Adapted from Schmitz J: Maldigestion and malabsorption. In Walker WA, Durie PR, Hamilton JR, et al, editors: Pediatric gastrointestinal disease, ed 3, Hamilton, Ontario, 2000, BC Decker, p 55.

330.1 Evaluation of Children with Suspected Intestinal Malabsorption

Michael J. Lentze and David Branski

The investigation is guided by the history and physical examination. In a child presenting with chronic or recurrent diarrhea, the initial work-up should include stool cultures and antibody tests for parasites, stool microscopy for ova and parasites such as Giardia, and stool occult blood and leukocytes to exclude inflammatory disorders. Stool pH and reducing substances for carbohydrate malabsorption, and quantitative stool fat examination and α₁-antitrypsin to demonstrate fat and protein malabsorption, respectively, should also be determined. Fecal stool elastase-1 can determine exocrine pancreatic insufficiency.

A complete blood count including peripheral smear for microcytic anemia, lymphopenia (lymphangiectasia), neutropenia (Shwachman syndrome), and acanthocytosis (abetalipoproteinemia) is useful. If celiac disease is suspected, serum immunoglobulin A (IgA) and tissue transglutaminase (TG2) antibody levels should be determined. Depending on the initial test results, more-specific investigations can be planned.

Investigations for Carbohydrate Malabsorption

Measurement of carbohydrate in the stool, using a Clinitest reagent that identifies reducing substances, is a simple screening test. An acidic stool with >2+ reducing substance suggests carbohydrate malabsorption. Sucrose or starch in the stool is not recognized as a reducing sugar until after hydrolysis with hydrochloric acid, which converts them to reducing sugars.

Breath hydrogen test is used to identify the specific carbohydrate that is malabsorbed. After an overnight fast, the suspected sugar (lactose, sucrose, fructose, or glucose) is administered as an oral solution (carbohydrate load 1-2 g/kg, maximum 50 g). In malabsorption, the sugar is not digested or absorbed in the small bowel, passes on to the colon, and is metabolized by the normal bacteria flora. One of the products of this process is hydrogen gas, which is absorbed through the colon mucosa and excreted in the breath. Increased hydrogen concentration in the breath samples suggests carbohydrate malabsorption. A rise in breath hydrogen of 20 ppm above the baseline is considered a positive test. The child should not be on antibiotics at the time of the test, because colonic flora is essential for fermenting the sugar.

Small bowel mucosal biopsies can measure mucosal disaccharidase (lactase, sucrase, maltase, palatinase) concentrations directly. In primary enzyme deficiencies the mucosal enzyme levels are low and small bowel mucosal morphology is normal. Partial or total villous atrophy due to disorders such as celiac disease, or following rotavirus gastroenteritis can result in secondary disaccharidase deficiency and transient lactose intolerance. The disaccharidase levels revert to normal after mucosal healing.

Investigations for Fat Malabsorption

The presence of fat globules in the stool suggests fat malabsorption. The ability to assimilate fat varies with age; a premature infant can absorb only 65-75% of dietary fat, a full-term infant absorbs almost 90%, and an older child absorbs >95% of fat while on a regular diet. Quantitative determination of fat malabsorption requires a 3-day stool collection for evaluation of fat excretion and determination of the coefficient of fat absorption:

where fat intake and fat losses are in grams. Because fecal fat balance studies are cumbersome, expensive, and unpleasant to perform, simpler tests are often preferred. Among these stool tests, the acid steatocrit test is the most reliable. When bile acid deficiency is suspected of being the cause of fat malabsorption, the evaluation of bile acid levels in duodenal fluid aspirate may be useful.

Fat malabsorption and exocrine pancreatic insufficiency are usually associated with deficiencies of fat-soluble vitamins A, D, E, and K. Serum concentrations of vitamins A, D, and E can be measured. A prolonged prothrombin time is an indirect test to assess vitamin K deficiency.

Investigations for Protein Losing Enteropathy

Dietary and endogenous proteins secreted into the bowel are almost completely absorbed; <1 g of protein from these sources passes into the colon. The majority of the stool nitrogen is derived from gut bacterial proteins. Excessive bowel protein loss usually manifests as hypoalbuminemia. However, the most common cause of hypoalbuminemia in children is a renal disorder; therefore, urinary protein excretion must be determined. Other potential causes of hypoalbuminemia include liver disease (reduced production) and inadequate protein intake. Very rarely hypoalbuminemia can result from an extensive skin disorder causing protein loss via the skin. Measurement of stool α₁-antitrypsin is a useful screening test for protein-losing enteropathy. This serum protein has a molecular weight similar to albumin’s; however, unlike albumin it is resistant to digestion in the gastrointestinal (GI) tract. Excessive α₁-antitrypsin excretion in the stool should prompt further investigations to identify the specific cause of gut or stomach (Menetrier disease) protein loss.

Investigations for Exocrine Pancreatic Function (Fig. 330-2)

Cystic fibrosis is the most common cause of exocrine pancreatic insufficiency in children; therefore, a sweat chloride test must be performed before embarking on invasive tests to investigate possible exocrine pancreatic insufficiency. Many cases of cystic fibrosis are detected by neonatal genetic screening programs; occasional rare mutations are undetected.

Figure 330-2 Algorithm for assessment of exocrine pancreatic function. *If not available, use other test. Perform appropriate imaging studies of the pancreas. **In case of borderline values, consider repeating the test with three independent samples. ***Consider differential diagnosis (especially consider mucosal villous atrophy and dilution effect of watery stool). GI, gastrointestinal.

(Adapted from Walkowiak J, Nousia-Arvanitakis S, Henker J, et al: Indirect pancreatic function tests in children, J Pediatr Gastroenterol Nutr 40:107–114, 2005.)

Fecal elastase-1 estimation is a sensitive test to assess exocrine pancreatic function in chronic cystic fibrosis and pancreatitis. Elastase-1 is a stable endoprotease unaffected by exogenous pancreatic enzymes. One disadvantage of the fecal elastase-1 test is the lack of full differentiation between primary exocrine pancreatic insufficiency and exocrine pancreatic dysfunction secondary to intestinal villous atrophy. The proximal small bowel is the site for pancreozymin/cholecystokinin production; the latter is the hormone that stimulates enzyme secretion from the exocrine pancreas. Mucosal atrophy can lead to diminished pancreozymin/cholecystokinin secretion and subsequently to exocrine pancreatic insufficiency. Fecal elastase-1 can also give a false-positive result during acute episodes of diarrhea.

Serum trypsinogen concentration can also be used as a screening test for exocrine pancreatic insufficiency. In cystic fibrosis, the levels are greatly elevated early in life, and then they gradually fall, so that by 5-7 yr of age, most patients with cystic fibrosis with pancreatic insufficiency have subnormal levels. Patients with cystic fibrosis and adequate exocrine pancreatic function tend to have normal or elevated levels. In such patients, observing the trend in serial serum trypsinogen estimation may be useful in monitoring exocrine pancreatic function. In Shwachman syndrome, another condition associated with exocrine pancreatic insufficiency, the serum trypsinogen level is low.

Other tests for pancreatic insufficiency (NBT-PABA test and pancreolauryl test) measure urine or breath concentrations of substances released and absorbed across the mucosal surface following pancreatic digestion. These tests lack specificity and are rarely used in clinical practice.

The gold standard test for exocrine pancreatic function is direct analysis of duodenal aspirate for volume, bicarbonate, trypsin and lipase upon secretin and pancreozymin/cholecysto-kinin stimulation. This involves duodenal intubation, and only a few centers perform this test (Chapter 340).

Investigations for Intestinal Mucosal Disorders

Establishing a specific diagnosis for malabsorption often requires histologic examination of small bowel mucosal biopsies. These are obtained during endoscopy, which allows multiple biopsies to be performed, because mucosal involvement can be patchy, especially in celiac disease. Periodic acid–Schiff (PAS) staining of mucosal biopsies and electron microscopy are necessary in congenital diarrhea to assess congenital microvillus atrophy. Bowel mucosal lesions can also be segmental in cases of intestinal lymphangiectasia. In these situations radiographic small bowel series or repeated ultrasonographies can identify a region of thickened bowel responsible for protein loss. During endoscopy, mucosal biopsies can be obtained to measure mucosal disaccharidase activities. Duodenal aspirates can be performed to measure pancreatic enzyme concentration as well as quantitative bacterial cultures. Aspirates to demonstrate other infections and infestations such as Giardia may be useful.

Imaging Procedures

Plain radiographs and barium contrast studies might suggest a site and cause of intestinal motility disorders. Although flocculations of barium and dilated bowel with thickened mucosal folds have been attributed to diffuse malabsorptive lesions such as celiac disease, these abnormalities are nonspecific. Diffuse fluid-filled bowel loops during sonography also suggest malabsorption.

330.2 Gluten-Sensitive Enteropathy (Celiac Disease)

David Branski, Riccardo Troncone

Etiology and Epidemiology

Celiac disease is an immune-mediated disorder elicited by the ingestion of gluten in genetically susceptible persons and characterized by chronic inflammation of the small intestine. It is considered an autoimmune condition because of the presence of anti–TG2 antibodies and the association with other autoimmune diseases (thyroid, liver, diabetes, adrenal).

Celiac disease is triggered by the ingestion of wheat gluten and related prolamines from rye and barley. In most studies oats proved to be safe; however, a few celiac patients have oats prolamine–reactive mucosal T cells that can cause mucosal inflammation.

Celiac disease is a common disorder (1% prevalence of biopsy-proven disease). It is thought to be rare in Central Africa and East Asia. Environmental factors might affect the risk of developing celiac disease or the timing of its presentation. Prolonged breastfeeding has been associated with a reduced incidence of symptomatic disease. Less clear is the effect of the time of gluten introduction in the infant diet; the ingestion of increased amounts of gluten in the 1st year of life can increase the incidence. Infectious agents have been hypothesized to play a role because frequent rotavirus infections are associated with an increased risk. It is plausible that the contact with gliadin at a time when there is ongoing intestinal inflammation, altered intestinal permeability, and enhanced antigen presentation can increase the risk of developing the disease, at least in a subset of persons (Fig. 330-3).

Figure 330-3 Causative factors in celiac disease. HLA, human leukocyte antigen.

(From Di Sabatino A, Corazza GR: Coeliac disease, Lancet 373:1480-1490, 2009.)

Genetics and Pathogenesis

A genetic predisposition is suggested by the family aggregation and the concordance in monozygotic twins, which approaches 100%. It is suggested that the primary association of CD is with the DQ αβ heterodimer encoded by the DQA1*05 and the DQB1*02 genes. Such a DQ molecule is present in ≥95% of celiac patients compared with 20-30% of controls. DQ2-negative celiac patients are invariably HLA DQ8 positive (DQA1*0301/DQB1*0302). A gene dosage effect has been suggested, and a molecular hypothesis for such a phenomenon has been proposed, based on the impact of the number and quality of the HLA DQ2 molecules on gluten peptide presentation to T cells. Other non-HLA genes confer susceptibility to celiac disease. Genome-wide association studies have shown risk variants in genes controlling the immune response, some being shared with type 1 diabetes.

Celiac disease is a T cell–mediated chronic inflammatory disorder with an autoimmune component. Altered processing by intraluminal enzymes, changes in intestinal permeability, and activation of innate immunity mechanisms may be involved and precede the activation of the adaptive immune response. Immunodominant epitopes from gliadin are highly resistant to intraluminal and mucosal digestion; incomplete degradation favors the immunostimulatory and toxic effects. Some gliadin peptides (p31-43) can activate innate immunity, and in particular they induce interleukin 15 (IL-15). Others activate lamina propria T cells in the context of HLA-DQ2 or DQ8 molecules. Gliadin-specific T-cell responses are enhanced by the action of TG2; the enzyme converts particular glutamine residues into glutamic acid, which results in higher affinity of these gliadin peptides for HLA-DQ2 or HLA-DQ8. The pattern of cytokines produced following gliadin activation is dominated by interferon-γ (IFN-γ) (Th1 skewed); IFN-α, IL-18, and IL-21 are also upregulated. A complex remodeling of the mucosa then takes place, involving increased levels of metalloproteinases and growth factors, which leads to the classic flat mucosa. Increased density of CD8+ cytotoxic intraepithelial lymphocytes are a hallmark of celiac disease. IL-15 is implicated in the expression of natural killer receptors CD94 and NKG2D, as well as in epithelial expression of stress molecules, thus enhancing cytotoxicity, cell apoptosis, and villous atrophy.

The most evident expression of autoimmunity is the presence of serum antibodies to TG2. However, the mechanisms leading to autoimmunity are largely unknown. The finding of IgA deposits on extracellular TG2 in the liver, lymph nodes, and muscles indicates that TG2 is accessible to the gut-derived autoantibodies. Several extraintestinal clinical manifestations of celiac disease (e.g., liver, heart, nervous system) are possibly related to the presence of autoantibodies.

Clinical Presentation and Associated Disorders

Clinical features of celiac disease vary considerably (Table 330-4). Intestinal symptoms are common in children whose disease is diagnosed within the 1st 2 years of life; failure to thrive, chronic diarrhea, vomiting, abdominal distention, muscle wasting, anorexia, and irritability are present in most cases (see Fig. 330-1). Occasionally there is constipation, rectal prolapse, or intussusception. As the age at presentation of the disease shifts to later in childhood, and with the more liberal use of serologic screening tests, extraintestinal manifestations and associated disorders, without any accompanying digestive symptoms, have increasingly become recognized, affecting almost all organs (Table 330-5).

Table 330-4 SOME CLINICAL MANIFESTATIONS OF CELIAC DISEASE IN CHILDREN AND ADOLESCENTS

SYSTEM	MANIFESTATION	(POSSIBLE) CAUSE
Gastrointestinal	Diarrhea Distended abdomen Vomiting Anorexia Weight loss Failure to thrive Aphthous stomatitis	Atrophy of the small bowel mucosa Malabsorption
Hematologic	Anemia	Iron malabsorption
Skeletal	Rickets Osteoporosis Enamel hypoplasia of the teeth	Calcium/vitamin D malabsorption
Muscular	Atrophy	Malnutrition
Neurologic	Peripheral neuropathy Epilepsy Irritability	Thiamine/vitamin B₁₂ deficiency
Endocrinologic	Short stature Pubertas tarda Secondary hyperparathyroidism	Malnutrition Calcium/vitamin D malabsorption
Dermatologic	Dermatitis herpetiformis Alopecia areata Erythema nodosum	Autoimmunity
Respiratory	Idiopathic pulmonary hemosiderosis

Adapted from Mearin ML: Celiac disease among children and adolescents, Curr Prob Pediatr Adolesc Health Care 37:81–112, 2007.

Table 330-5 RISK GROUPS FOR CELIAC DISEASE CASE-FINDING

1st-degree relatives

Dermatitis herpetiformis

Unexplained iron deficiency anaemia

Autoimmune thyroiditis

Type 1 diabetes

Unexplained infertility

Recurrent abortion

Dental enamel hypoplasia

Cryptic hypertransaminasemia

Autoimmune liver disease

Short stature

Delayed puberty

Down, Williams, and Turner syndromes

Irritable bowel syndrome

Unexplained osteoporosis

Sjögren syndrome

Epilepsy with occipital calcifications

Selective IgA deficiency

Addison disease

IgA, immunoglobulin A.

Modified from Di Sabatino A, Corazza GR: Coeliac disease, Lancet 373:1480–1490, 2009.

The most common extraintestinal manifestation of celiac disease is iron-deficiency anemia, unresponsive to iron therapy. Osteoporosis may be present; in contrast to the situation in adults, it can be reversed by a gluten-free diet, with restoration of normal peak bone densitometric values. Other extraintestinal manifestations include short stature, endocrinopathies, arthritis and arthralgia, epilepsy with bilateral occipital calcifications, peripheral neuropathies, cardiomyopathy, chronic lung disease, isolated hypertransaminasemia, dental enamel hypoplasia, aphthous stomatitis, and alopecia. The mechanisms responsible for the severity and the variety of clinical presentations remain obscure. Nutritional deficiencies or abnormal immune responses have been advocated.

Silent celiac disease is being increasingly recognized, mainly in asymptomatic 1st-degree relatives of celiac patients investigated during screening studies. However, small bowel biopsy in these people reveals severe mucosal damage consistent with celiac disease. Potential celiac disease is defined when patients are identified by positive screening studies but without documented celiac disease on small bowel biopsy. It is important to follow these patients because they can develop established celiac disease in the future (Table 330-6).

Table 330-6 CLINICAL SPECTRUM OF CELIAC DISEASE

SYMPTOMATIC

Frank malabsorption symptoms: chronic diarrhea, failure to thrive, weight loss

Extraintestinal manifestations: anemia, fatigue, hypertransaminasemia, neurologic disorders, short stature, dental enamel defects, arthralgia, aphthous stomatitis

SILENT

No apparent symptoms in spite of histologic evidence of villous atrophy

In most cases identified by serologic screening in at-risk groups (see Table 330-1)

LATENT

Subjects who have a normal histology, but at some other time, before or after, have shown a gluten-dependent enteropathy

POTENTIAL

Subjects with positive celiac disease serology but without evidence of altered jejunal histology

It might or might not be symptomatic

Some diseases, many with an autoimmune pathogenesis, are found with a higher than normal incidence in celiac patients. Among these are type 1 diabetes, autoimmune thyroid disease, Addison disease, Sjögren syndrome, autoimmune cholangitis, autoimmune hepatitis, primary biliary cirrhosis, IgA nephropathy, alopecia, and dilated cardiomyopathy. Such associations have been interpreted as a consequence of the sharing of identical HLA haplotypes. The relation between celiac disease and other autoimmune diseases is poorly defined; once those diseases are established, they are not influenced by a gluten-free diet. Other associated conditions include selective IgA deficiency, Down syndrome, Turner syndrome, and Williams syndrome.

Patients with celiac disease show increased long-term mortality, the risk rising with delayed diagnosis and/or poor dietary compliance. Non-Hodgkin lymphoma is the main cause of death. Adult patients can develop complications such a refractory celiac disease, ulcerative jejunoileitis, or enteropathy-associated T-cell lymphoma.

Diagnosis

Serologic tests have a crucial role in the diagnosis of celiac disease; sensitivity of the IgA anti-TG2 is 61-100% (mean, 87%), and specificity is 86-100% (mean, 95%). Some 10% of patients whose disease is diagnosed earlier than 2 yr of age show absence of IgA anti-TG2. For them, the measurement of serum antigliadin antibodies is generally advised. Antibodies against gliadin-derived deamidated peptides (D-AGA) have been assessed. Compared with conventional AGA, the peptide antibodies (IgG and IgA) have a greater sensitivity and specificity. A problem with serology is represented by the association of celiac disease with IgA deficiency (10-fold increase compared to the general population). Serum IgA should always be checked, and in the case of IgA deficiency, D-AGA, IgG anti-endomysium, or TG2 should be sought. Negative serology should not preclude a biopsy examination when the clinical suspicion is strong.

Genetic tests have an increasing role in the diagnosis. Less than 2% of celiac patients lack both HLA specificities; at the same time, approximately one third of the “normal” population has one or the other marker; that means that the measurement of HLA DQ2 and/or DQ8 has a strong negative predictive value but a very weak positive predictive value for the diagnosis of celiac disease. With these limitations the test can prove useful to exclude celiac disease when the genetic studies are negative in subjects on a gluten-free diet or in subjects belonging to an at-risk group (e.g., 1st-degree relatives, insulin-dependent diabetics, patients with Down syndrome) to avoid long-term follow-up.

The ultimate diagnosis of celiac disease relies on the demonstration of specific, though not pathognomonic, histopathologic abnormalities in the small bowel mucosa (Table 330-7). According to The European Society for Pediatric Gastroenterology, Hepatology and Nutrition (ESPGHAN) current criteria, the 2 requirements mandatory for the diagnosis of celiac disease are the finding of villous atrophy with hyperplasia of the crypts and abnormal surface epithelium, while the patient is eating adequate amounts of gluten, and a full clinical remission after withdrawal of gluten from the diet. The finding of circulating IgA celiac disease–associated antibodies at the time of diagnosis and their disappearance on a gluten-free diet adds weight to the diagnosis. A control biopsy to verify the consequences of the gluten-free diet on the mucosal architecture is considered mandatory only in patients with an equivocal clinical response to the diet. Gluten challenge is not considered mandatory except in situations where there is doubt about the initial diagnosis, for example, when an initial biopsy was not performed or when the biopsy specimen was inadequate or atypical of celiac disease.

It is now accepted that the spectrum of histologic abnormalities in the celiac small intestine is wider than previously recognized. In some celiac disease patients, only subtle changes of crypt elongation with an increase in intraepithelial lymphocytes may be present. In those cases, it is very important to also evaluate the serology and the HLA typing so as to reach the correct diagnosis. Analysis of multiple biopsies is also very important.

Nevertheless, many cases of celiac disease are undiagnosed, and the ratio between patients with diagnosed and with undiagnosed disease may be as high as 1 : 7. Case finding by liberal use of anti-endomysium or anti-TG2 antibodies, followed by confirmatory jejunal biopsy, is more cost effective in primary care than mass screening is. Patients with symptoms or diseases known to be associated with celiac disease should undergo serologic evaluation.

Treatment

The only treatment for celiac disease is lifelong strict adherence to a gluten-free diet (Fig. 330-4). This requires a wheat-, barley-, and rye-free diet. Despite evidence that oats are safe for most patients with celiac disease, there is concern regarding the possibility of contamination of oats with gluten during harvesting, milling, and shipping. Nevertheless, it seems wise to add oats to the gluten-free diet only when the latter is well established, so that possible adverse reactions can be readily identified. There is a consensus that all celiac disease patients should be treated with a gluten-free diet regardless of the presence of symptoms. However, whereas it is relatively easy to assess the health improvement after treatment of celiac disease in patients with clinical symptoms of the disease, it proves difficult in persons with asymptomatic celiac disease. The nutritional risks, particularly osteopenia, are those mainly feared for subjects who have silent celiac disease and continue on a gluten-containing diet. Little is known about the health risks in untreated patients with minor enteropathy, which may be clinically silent. There are no guidelines concerning the need for a gluten-free diet in subjects with “potential” celiac disease (patients with positive celiac disease–associated serology but without enteropathy).