INTRODUCTION — Celiac disease, also known as gluten-sensitive enteropathy, is a common immune-mediated inflammatory disease of the small intestine caused by sensitivity to dietary gluten and related proteins in genetically predisposed individuals. It differs from food allergies (including wheat allergy), which are mediated by immunoglobulin E (IgE) or immunoglobulin G (IgG). The appropriate treatment is a gluten-free diet for life, which results in complete resolution of symptoms and mucosal healing in most individuals and probably also reduces the long-term risk for adverse complications including intestinal lymphoma.
The epidemiology, pathogenesis, and clinical manifestations of celiac disease are reviewed here. Other aspects of celiac disease in children are discussed in the following topic reviews:
●(See "Diagnosis of celiac disease in children".)
●(See "Management of celiac disease in children".)
EPIDEMIOLOGY
General population — Celiac disease is a common chronic condition and is estimated to affect approximately 0.5 to 1 percent of the general population in many parts of the world. In Europe, the United States, and Australia, prevalence estimates range from 1:80 to 1:300 children (3 to 13 per 1000 children) [1]. Females are affected approximately twice as often as males, although the ratio varies depending on the strategy used to find cases [2]. The number of so-called "silent celiacs" (a misnomer because most of these patients suffer from nonspecific symptoms) is much higher than the number of patients with classic celiac disease. As an example, a study from Italy reported that asymptomatic cases outnumbered symptomatic cases by a ratio of 7:1 (table 1) [3]. A large serologic screening study in the United States suggested a prevalence of 1:133 among patients with no risk factors or symptoms [4].
Celiac disease is also common in some non-European populations, including northern Africa, the Middle East, and South Asia [1]. The prevalence in South America varies considerably by country [1,5,6]. Previously thought to be uncommon in China, a few cases have now been reported [7], and a report based on serologic testing among adolescents and young adults in China suggests that the prevalence may be as high as 0.76 percent [8]. The prevalence in some developing countries is probably underestimated due to limited access to diagnostic testing and confounding by other diarrheal diseases that cause small intestinal injury. Overall, the global distribution of the disease seems to parallel the distribution of human leukocyte antigen (HLA) genotypes that predispose to celiac disease, provided that the population is also exposed to gluten [9].
High-risk groups — The prevalence of celiac disease as detected by screening programs using specific antibodies is substantially increased in the following groups as compared with the general population (table 2). Because of the increased risk, we suggest routine screening for celiac disease in asymptomatic children with these conditions, although this approach is somewhat controversial (see "Diagnosis of celiac disease in children", section on 'Members of high-risk groups'):
●First- and second-degree relatives of patients with celiac disease [4,10]
●Down syndrome
●Type 1 diabetes
●Selective immunoglobulin A (IgA) deficiency
●Autoimmune thyroiditis
●Turner syndrome
●Williams syndrome
●Juvenile chronic arthritis
For these groups, the risk of celiac disease is approximately 3- to 10-fold higher than in the general population [4,10-16]. Evidence for these associations is discussed below.
Relatives of patients with celiac disease — For first-degree relatives of patients with celiac disease, the prevalence is approximately 7.5 percent (1:13) and for second-degree relatives, approximately 2.3 percent (1:43) [17]. The prevalence varies with the relationship to the proband. In a meta-analysis, the pooled prevalence of celiac disease was 8.9 percent (1:11) for siblings of the proband, 7.9 percent (1:13) for offspring of the proband, and 3 percent (1:33) for parents of the proband [17]. The risk was higher in female relatives of the proband compared with male relatives.
Down syndrome — Celiac disease occurs in 5 to 12 percent of individuals with Down syndrome [18-22]. Pooled prevalence estimates for celiac disease are 1:13 (7.6 percent), approximately a fivefold increase in risk over the general population [19]. The mechanism for this association is unclear, although it is consistent with the propensity for other autoimmune diseases among individuals with Down syndrome. (See "Down syndrome: Clinical features and diagnosis", section on 'Gastrointestinal abnormalities'.)
Type 1 diabetes mellitus — Celiac disease is associated closely with type 1 diabetes mellitus [13,23-25]. In several reports, between 2.6 and 7.8 percent of adults with type 1 diabetes had IgA autoantibodies to endomysium or to tissue transglutaminase (tTG); most such patients were proven to have celiac disease with small bowel biopsy [19,26,27]. Many such patients had no overt clinical manifestations of celiac disease [26]. Other reports have identified celiac disease in as many as 3.5 percent of children whose parents have type 1 diabetes, the prevalence of which increases with age [12]. Of note, celiac disease and type 1 diabetes share HLA susceptibility alleles [19], but the co-occurrence of the disorders is not fully explained by shared genetic risk loci [25]. (See "Associated autoimmune diseases in children and adolescents with type 1 diabetes mellitus", section on 'Celiac disease'.)
A causal relationship between celiac disease and diabetes mellitus has been suggested but not established. A few studies in humans [28] and animals [29,30] suggest that celiac disease may trigger autoimmune processes leading to diabetes. One study noted that the prevalence of autoimmune diseases, including type 1 diabetes mellitus, may be related to the duration of exposure to gluten and may reach more than 30 percent in patients diagnosed with celiac disease after age 20 [31]. However, other observations suggest that celiac disease does not trigger diabetes, since the age of onset and the severity of diabetes do not appear to be influenced by the presence of celiac disease [26] and celiac autoantibodies usually develop after the onset of diabetes [25,32]. Thus, larger and prospective clinical studies are required to clarify the relationship between celiac disease, type 1 diabetes, and other autoimmune disorders.
Selective IgA deficiency — Approximately 8 percent of individuals with selective IgA deficiency have celiac disease [33]; conversely, 1.7 to 3 percent of individuals with celiac disease have IgA deficiency [19]. Because most specific serologic tests for celiac disease are IgA antibodies, patients with known IgA deficiency require special strategies when screening for celiac disease. In addition, the possibility of undiagnosed IgA deficiency should be considered when interpreting the results of serologic screens. (See "Diagnosis of celiac disease in children", section on 'Special populations'.)
Autoimmune thyroiditis — Approximately 2 to 7 percent of individuals with autoimmune thyroiditis develop celiac disease [19,34-36]. The association is weak during childhood and appears to increase with age [10]. Conversely, 10 to 20 percent of individuals with celiac disease have autoimmune thyroiditis, and its clinical course does not appear to be affected by a gluten-free diet [37,38]. (See "Acquired hypothyroidism in childhood and adolescence", section on 'Autoimmune thyroiditis'.)
Despite this association, the presence of antithyroid antibodies at diagnosis has a low predictive value for the development of thyroid hypofunction. In a series of 135 children with celiac disease, 23 percent had positive antithyroid antibody titers at diagnosis [39]. Approximately 12 percent of the patients had elevated thyroid-stimulating hormone (TSH) levels, suggesting subclinical hypothyroidism, regardless of whether the antithyroid antibody titers were positive. In most of these patients, the subclinical hypothyroidism normalized during the follow-up period (8.9±4 years on a gluten-free diet). Among patients with persistently positive antithyroid antibodies during the follow-up period, subclinical hypothyroidism developed in 25 percent. Thus, this study suggests that treatment with a gluten-free diet does not alter the clinical course of autoimmune thyroiditis in patients with celiac disease.
Other associations — The prevalence of celiac disease is also increased in Turner syndrome [10,14,40]. A systematic review reported that a pooled prevalence estimate for celiac disease in Turner syndrome is 6.3 percent; it is unclear whether the HLA susceptibility alleles are increased in patients with Turner syndrome [19]. Patients with Williams syndrome also may have an increased prevalence of celiac disease (up to 9.5 percent), but the available data are limited [15,41]. (See "Clinical manifestations and diagnosis of Turner syndrome", section on 'Autoimmune disorders' and "Williams syndrome".)
One study reported that 2 percent of children evaluated for rheumatic complaints had celiac disease (approximately threefold the rate in the general population) [42]. (See 'Arthritis' below.)
Weaker associations with primary biliary cholangitis, as well as a variety of other liver diseases, have been described in adults. Evidence for these associations is described separately. (See "Epidemiology, pathogenesis, and clinical manifestations of celiac disease in adults", section on 'Associated conditions'.)
PATHOGENESIS AND RISK FACTORS — Celiac disease is an immune-mediated inflammatory disease of the small intestine caused by sensitivity to dietary gluten and related proteins in genetically predisposed individuals. The grains that contain the triggering proteins are wheat, barley, and rye. Ingestion of these proteins by a susceptible individual causes immune-mediated mucosal inflammation of the proximal small intestine, with villous atrophy and crypt hyperplasia (picture 1A-B), which often leads to malabsorption and gastrointestinal symptoms. The immune response is mediated by gliadin-reactive T cells, in contrast with wheat allergy, which is mediated by IgE or IgG. The intestinal lesions and symptoms resolve when gluten is eliminated from the diet. (See "Epidemiology, pathogenesis, and clinical manifestations of celiac disease in adults", section on 'Pathogenesis' and "Management of celiac disease in children", section on 'Principles of a gluten-free diet'.)
Genetic factors — The genetic basis of the disease is shown by the frequent intrafamilial occurrence and the remarkably close association with the human leukocyte antigen (HLA) DR3-DQ2 and/or DR4-DQ8 gene locus. More than 99 percent of individuals with celiac disease have HLA DR3-DQ2 and/or DR4-DQ8, compared with approximately 40 percent of the general population [43]. Homozygotes for DR3-DQ2 are at the highest risk for celiac disease, which develops in approximately 10 percent of individuals with this genotype by five years of age [44]. While the presence of either the HLA-DQ2 or DQ8 genotype is essential to confer disease, it is not sufficient, and other genes at non-HLA loci must also participate, in addition to environmental factors [1].
Because of common genetic contributors, several groups are at increased risk for celiac disease. The genetic contributors to celiac disease are discussed in detail separately. (See 'High-risk groups' above and "Epidemiology, pathogenesis, and clinical manifestations of celiac disease in adults", section on 'Genetic factors'.)
Autoimmunity — Celiac disease is associated with a number of autoimmune disorders, including type 1 diabetes mellitus and autoimmune thyroid disease. In addition, the intestinal lesion of celiac disease is associated with several different autoantibodies that are useful for diagnostic purposes. IgA-antibodies against endomysium and the endomysial autoantigen tissue transglutaminase (tTG) are highly sensitive and specific. (See 'High-risk groups' above and "Diagnosis of celiac disease in children", section on 'Pretesting diet'.)
While the presence of these antibodies is recognized, their precise role in the pathogenesis of celiac disease remains to be clarified. (See "Diagnosis of celiac disease in adults", section on 'Serum antibody assays'.)
Feeding practices in infancy and early childhood — The pathogenesis of celiac disease at any age requires exposure to gluten. Efforts to reduce the risk of celiac disease have focused on the initial timing and amount of gluten exposure and whether breastfeeding or the use of hydrolyzed formula (avoidance of cow's milk protein) reduce the risk of developing celiac disease.
●Breastfeeding and gluten exposure – The timing of initial gluten exposure during infancy and association with breastfeeding does not affect the risk for clinical expression of celiac disease. There is some evidence that the quantity of gluten in the infant's diet affects the age of onset of disease in genetically predisposed individuals. Current recommendations are to introduce gluten to an infant's diet any time between 4 and 12 months of age, as recommended in a position paper from the European Society for Paediatric Gastroenterology, Hepatology and Nutrition (ESPGHAN) [45,46]. This position paper also suggested against rapid introduction of large amounts of gluten, although this was not precisely defined. (See "Introducing solid foods and vitamin and mineral supplementation during infancy", section on 'What to feed and how to advance'.)
•Timing of gluten introduction – The possibility that the timing of initial gluten exposure concomitant with breastfeeding affects the risk for clinical expression of celiac disease was raised by observational studies [47-50], including an epidemic of celiac disease in Sweden during the 1990s that appeared to be related to changes in infant feeding practices [44,51,52]. However, two randomized trials failed to confirm this hypothesis and reported no association between the timing of gluten introduction or breastfeeding and celiac disease risk. One study included 944 infants in eight European countries who were at high risk for celiac disease because of their genetic profile (HLD-DQ-2 or DQ-8 positive and at least one first-degree relative with celiac disease). They were randomized to receive a small amount of gluten daily from 16 to 24 weeks of age and were compared with infants given placebo [53]. There was no difference in celiac disease prevalence at three years of age, and the prevalence did not vary with ongoing breastfeeding. A second study included 832 infants from Italy with a similarly high risk for celiac disease who were randomized to dietary introduction of gluten at six months of age versus 12 months of age [54]. Early introduction of gluten was associated with higher risk for celiac autoimmunity and overt celiac disease at two years of age, but there was no difference between the groups by five years of age and no association with concurrent breastfeeding. Thus, later introduction of gluten did not alter overall celiac disease risk, although it modestly delayed the onset of celiac disease.
•Quantity of gluten exposure – The quantity of gluten in the infant's diet may affect the risk for clinical expression of celiac disease or at least the earlier timing of its onset. In a prospective observational multinational study of more than 6605 children with genetic predisposition for celiac disease due to their HLA antigen genotype, the quantity of gluten exposure during the first five years of life was associated with development of celiac autoimmunity and confirmed celiac disease [55]. By three years of age, the absolute risk for developing celiac autoimmunity and celiac disease was 28 and 21 percent, respectively, among children who consumed the reference amount of gluten (mean intake 3.7 g/day), compared with 34 and 28 percent for those who consumed an additional 1 g/day of gluten. One g of gluten corresponds to approximately one-half slice of white bread or 75 g of cooked pasta. The high prevalence of celiac autoimmunity and celiac disease in this cohort reflects the genetic predisposition of the study population.
These findings are consistent with those of a previous study that found at least a twofold increase in risk for children with higher levels of gluten exposure up to two years of age [56], although a separate study found that the amount of gluten consumed between 11 and 36 months of age did not influence the risk of developing celiac disease [57]. Long-term follow-up of these children is needed to determine whether the amount of gluten consumed at an early age ultimately has any effect on the later development of celiac disease.
Contrasting results were reported from a subsequent study, which concluded that introduction of relatively large quantities of gluten into the diet between four and six months of age reduced the prevalence of celiac disease at three years of age [58]. These findings need to be confirmed in larger numbers of at-risk children before any firm recommendations in this regard can be made.
●Cow's milk protein – Avoidance of cow's milk protein during the weaning period does not appear to reduce the risk for celiac disease autoimmunity or celiac disease. This was shown in a randomized trial of 230 infants who were considered at risk for celiac disease because of their HLA type and at least one family member having type 1 diabetes [59]. Infants weaned to an extensively hydrolyzed formula did not have decreased risk for tTG positivity or celiac disease compared with those weaned to cow's milk-based formula, with up to 10 years follow-up. Individuals who developed celiac disease had increased titers of cow's milk antibodies prior to elevation of tTG antibodies or diagnosis of celiac disease, but there was no evidence that this was causal.
Additional trigger factors — In some cases, celiac disease in genetically predisposed individuals may be precipitated by a separate risk or trigger factor in addition to gluten exposure. Prior intestinal infections, and, in particular, infection with rotavirus or enterovirus, are more common in children with celiac disease than those without [60,61]. If true, the precise mechanisms by which these events affect the onset of disease remain unclear. Exposure to antibiotics is probably not a risk factor [62,63].
CLINICAL MANIFESTATIONS — In the past, celiac disease usually presented in infants and young children with malabsorption and failure to thrive. Now, celiac disease tends to present later, between the ages of 10 and 40, with milder gastrointestinal or non-gastrointestinal manifestations [64]. This changing presentation of the disease may be due to the later introduction of gluten into the infant diet and increasing recognition of subclinical disease due in part to advances in serologic screening [65].
"Classical" gastrointestinal symptoms — Classically, celiac disease presented between 6 and 24 months of age, after the introduction of gluten into the diet [10]. The children have chronic diarrhea, anorexia, abdominal distension and pain, and failure to thrive or weight loss; some may also have vomiting. If the diagnosis is delayed, children may present with signs of severe malnutrition. Severely affected infants may present with a celiac crisis and the hemodynamic and metabolic consequences of dehydration.
Gastrointestinal symptoms in older children and adults are similar but usually less dramatic. Paradoxically, the disease may cause either constipation (8 percent) or diarrhea (64 percent) [66]. When diarrhea is present, the stools are often bulky and foul-smelling and may float because of trapped air. Flatulence and abdominal distension (caused by colonic bacterial digestion of malabsorbed nutrients) are common. These symptoms may be accompanied by the consequences of malabsorption, such as weight loss, growth failure, severe anemia, neurologic disorders from deficiencies of B vitamins, and osteopenia from deficiencies of vitamin D and calcium.
Non-gastrointestinal manifestations — Numerous non-gastrointestinal manifestations of celiac disease have been described (table 3A). Conditions associated with celiac disease in adults are described in detail separately [67]. In some patients, non-gastrointestinal symptoms are the primary presenting complaint and should prompt the consideration of serologic testing. (See 'Non-gastrointestinal celiac disease' below and "Diagnosis of celiac disease in children" and "Epidemiology, pathogenesis, and clinical manifestations of celiac disease in adults".)
Growth and development — Children who develop symptomatic celiac disease often have delayed linear growth. Indeed, between 8 and 10 percent of children with apparent "idiopathic" short stature have serologic evidence of celiac disease [10]. The growth delay may occur even when weight-for-height is relatively normal and in the absence of significant gastrointestinal symptoms [68-70]. Thus, the growth attenuation is probably not entirely attributable to undernutrition. On the other hand, young children with persistently positive celiac-specific antibodies but without symptoms generally have normal growth, suggesting that the effects of celiac disease on growth probably depend on disease severity and/or duration [71].
Studies from the United States, Europe, and Australia report that a substantial number of children are overweight or obese at the time that celiac disease is diagnosed, although the rate of obesity is somewhat lower than in the local population [72-75]. In one series, 75 percent of patients who were overweight or obese at diagnosis had a decrease in their degree of obesity (decrease in body mass index Z-score) after beginning a gluten-free diet [72]. Among patients who were underweight at diagnosis, 80 percent grew into a healthy weight after beginning a gluten-free diet and 20 percent remained underweight.
Adolescent girls may have an increased frequency of menstrual abnormalities such as delayed menarche and, later, may have problems with infertility and experience an early menopause [76-80]. Boys with untreated celiac disease have reduced levels of serum dihydrotestosterone in a pattern suggesting androgen resistance [10,76]. Treatment with a gluten-free diet appears to prevent these problems. (See "Epidemiology, pathogenesis, and clinical manifestations of celiac disease in adults", section on 'Menstrual and reproductive issues'.)
Neurologic disease and behavioral symptoms — Neurologic or behavioral symptoms may be associated with celiac disease and occasionally are the primary symptoms.
●Short-term – Subclinical or subtle neurologic or behavioral abnormalities appear to be common and tend to resolve on a gluten-free diet. These may involve the central or peripheral nervous systems:
•Subtle behavioral abnormalities were demonstrated in a study of 3.5-year-old children, in which those with persistently positive tissue transglutaminase (tTG) antibodies on a gluten-containing diet were more likely to manifest behavioral symptoms (anxiety, depression, aggressive behavior, or sleep problems) compared with those with negative antibody tests [81]. Unlike previous studies, the parents were unaware of the child's elevated tTG at the time that they reported the symptoms, avoiding that potential bias. This finding was no longer observed at 4.5 years of age. A similar screening study in school-aged children also reported a modest association between tTG antibody positivity and anxiety symptoms [82]. The association may be explained by residual confounding, as with any observational study.
•Subclinical neurologic abnormalities were suggested by a study of children with newly diagnosed celiac disease, in which almost 20 percent had subclinical neurologic abnormalities [83]. Among 27 children, two had peripheral polyneuropathy documented with electromyography, one had prolonged latencies in somatosensory evoked potential, and two had central magnetic resonance imaging (MRI) abnormalities consisting of pontine demyelination or cortical atrophy. Similarly, there is some evidence of regional brain hypoperfusion in adults with untreated celiac disease [84]. In most, but not all, such conditions, improvement is observed after treatment with a gluten-free diet [85-88].
●Long-term – Celiac disease may be associated with a modestly increased risk of neurobehavioral disorders in long-term follow-up. The best evidence comes from a large case-control study in Sweden, which found that individuals with childhood-onset celiac disease had a 1.4-fold increased risk of developing a neurobehavioral or psychiatric disorder in long-term follow-up (median follow-up 9.6 years), including mood, anxiety, eating, and behavioral disorders; attention deficit hyperactivity disorder (ADHD); autism spectrum disorder; and intellectual disability [89]. This higher risk was not observed in siblings of the patients with celiac disease, arguing against a genetic component. Other less robust studies also suggest an association between celiac disease and mood disorders, behavioral disorders, developmental delay, ADHD, headache, and cerebellar ataxia [86,90-92].
Epileptic disorders are only slightly more common among children with celiac disease, and there is no increase in frequency of tic disorders. Conversely, celiac disease does not appear to be overrepresented in groups of children with neurologic or psychiatric disorders [91,93].
Reports of neurologic and neuropsychiatric symptoms in adults with celiac disease are discussed separately. (See "Epidemiology, pathogenesis, and clinical manifestations of celiac disease in adults", section on 'Neuropsychiatric'.)
The pathogenesis of the neurologic symptoms is unclear. Some of the disorders, such as infantile hypotonia and developmental delay, may be caused by malnutrition, including specific micronutrient deficiencies; these problems tend to resolve on a gluten-free diet. However, there is increasing evidence that some or all of these neurologic abnormalities are caused by autoimmune mechanisms. As an example, widespread tTG-IgA deposition around blood vessels in the cerebellum has been described [94]. In particular, antiganglioside antibodies may be involved in the pathogenesis of neurologic symptoms [95], although relevant studies have yielded somewhat conflicting results. These findings suggest that an immune-mediated process may lead to gluten ataxia and/or peripheral neuropathy [94].
Liver disease — Studies of children with celiac disease suggest that aminotransferase elevations are also common at diagnosis (15 to 35 percent), particularly in patients presenting with the classical symptoms of the disease [96-98]. In most patients, the aminotransferases normalize with a gluten-free diet. Similarly, mild elevations in serum alanine aminotransferase (ALT) or aspartate aminotransferase (AST) were seen in 42 percent of adult patients with celiac disease [99]. Conversely, celiac disease is found in 5 to 10 percent of adults with chronic elevations of aminotransferases [100].
Patients with celiac disease also appear to have increased risks for a broad spectrum of liver diseases, including acute hepatitis, primary biliary cholangitis (previously known as primary biliary cirrhosis), and chronic hepatitis including autoimmune hepatitis [101-104]. Several cases of severe liver disease with cirrhosis in children with celiac disease have been reported [105], but celiac disease was not established as a causative factor. (See "Epidemiology, pathogenesis, and clinical manifestations of celiac disease in adults", section on 'Liver disease'.)
Iron deficiency — The prevalence of iron deficiency is increased among children with celiac disease and appears to be correlated with severity of mucosal injury. In one prospective study, the prevalence of low ferritin (defined in this case as <10 micrograms/mL) was 21 percent in patients with "potential" celiac disease, 35 percent in those with partial villous atrophy, and 87 percent among those with total villous atrophy, compared with 0 percent in healthy controls [106]. For this reason, testing for celiac disease should be considered in children with iron deficiency anemia who have no other clear reason for the deficiency.
Dermatitis herpetiformis — There are a number of skin manifestations of celiac disease (table 3B). Dermatitis herpetiformis is the most common, occurring in up to 24 percent of adult patients with celiac disease [26,107]. It is rare in children, but cases have been reported [108,109]. Similar to celiac disease, antibodies to tissue transglutaminase (anti-tTG) are elevated in patients with dermatitis herpetiformis, confirming the pathogenic similarities between these diseases [110].
Approximately 85 percent of adult patients with dermatitis herpetiformis have the characteristic changes of celiac disease on intestinal biopsy, although most have no gastrointestinal symptoms. Dermatitis herpetiformis is less common prior to puberty but has been reported in patients as young as eight months old [111,112].
Dermatitis herpetiformis is characterized by an itchy papular vesicular eruption, usually located symmetrically on the extensor surfaces of the elbows, knees, buttocks, sacrum, face, neck, trunk, and, occasionally, within the mouth (picture 2A-B). It is commonly misdiagnosed as atopic dermatitis, scabies, or linear IgA dermatosis [113]. The clinical manifestations and diagnosis of dermatitis herpetiformis are discussed separately. (See "Dermatitis herpetiformis", section on 'Clinical findings'.)
Many experts recommend a lifelong gluten-free diet based on the results of the skin biopsy alone, and an intestinal biopsy is not required. Although patients with dermatitis herpetiformis may have a symptomatic response to medications such as dapsone, the skin lesions usually will not resolve without gluten withdrawal [114]. (See "Dermatitis herpetiformis", section on 'Treatment'.)
Dental enamel defects — Dental enamel defects involving the secondary dentition are more common among children and adults with celiac disease compared with the general population and may occur in the absence of gastrointestinal symptoms [115-117]. The enamel defects considered to be specific to celiac disease are symmetrically distributed and detectable in all four quadrants of the dentition [116]. Defects may consist of cream, yellow, or brown opacities; loss of enamel glaze; horizontal grooves; or shallow pits (picture 3). The incisors are most commonly affected. There is some evidence that these defects are mediated by immunologic mechanisms (associated with the human leukocyte antigen [HLA] allele DR3) and not by malabsorption of nutrients such as calcium [118]. Early identification and treatment of celiac disease may prevent the development of the enamel defects [119].
Metabolic bone disease — Bone loss (usually osteomalacia) occurs commonly in celiac disease and can occur in patients without gastrointestinal symptoms [69,120-124]. These patients have secondary hyperparathyroidism that is probably caused by vitamin D deficiency [125,126].
In children, metabolic bone disease generally resolves with a gluten-free diet [124,126-128]. In a study of 30 children and adolescents maintained on a long-term gluten-free diet (average 10.7 years), bone mineral density and serum markers of bone metabolism completely normalized [128].
Arthritis — Approximately 25 percent of adults with celiac disease have arthritis [129], but the prevalence of arthritis in children with celiac disease is not well established. Conversely, celiac disease is reported in 2 to 3 percent of children presenting with juvenile idiopathic arthritis or juvenile chronic arthritis [42,130,131]. In one large cohort of 2125 children evaluated for rheumatic complaints, 2 percent were found to have celiac disease after routine screening, the majority of whom had no gastrointestinal symptoms [42]. In all cases, the musculoskeletal symptoms (primarily myalgias and arthralgias) resolved after beginning a gluten-free diet.
Subclinical disease — The development and widespread availability of serologic screening has led to the understanding that celiac disease can exist in a very mild form and may go largely undetected because most patients have mild and nonspecific symptoms, such as fatigue, borderline iron deficiency or anemia, attenuated growth, or otherwise unexplained elevations in serum aminotransferases [69,132,133] or no symptoms at all [123]. The most common type of anemia in celiac disease is caused by iron deficiency; megaloblastic anemia is rare. Serum iron, serum folate, and red cell folate are sometimes reduced in patients with untreated celiac disease who are older than one year. (See 'Iron deficiency' above.)
The range of symptoms in children with subclinical disease is illustrated by a study of children whose celiac disease was diagnosed through a screening program [3]. Most of these children had minimal gastrointestinal symptoms. However, there were numerous important clinical and laboratory findings, such as iron deficiency, recurrent abdominal pain, and mood changes (table 4). In another study, 31 percent of patients with subclinical disease (versus 67 percent with classic symptoms) were malnourished (defined as body weight less than 90 percent of ideal) [134]. Once on a gluten-free diet, all reported objective and subjective improvement of well-being as they recognized symptoms they had not previously considered to be abnormal, especially fatigue and abdominal pain.
Even in individuals with minimal symptoms, establishing and treating subclinical celiac disease may help to identify and treat unsuspected nutritional deficiencies and to reduce the risk of low birth weight infants born to affected mothers. It is less clear whether these individuals have increased risk for autoimmune diseases or malignancies that might be reduced by treatment with a gluten-free diet.
Risk of malignancy — Several reports have suggested increased risk for some malignancies, particularly non-Hodgkin lymphoma and gastrointestinal cancers, in adults with celiac disease compared with the general population. The incidence of cancer does not appear to be increased during childhood or adolescence.
A few studies suggest that the risk for malignancy is reduced by long-term treatment with a gluten-free diet. Although this has not been fully established, it is one of the rationales for recommending lifelong treatment for all patients with celiac disease, even for those with minimal gastrointestinal symptoms. (See "Epidemiology, pathogenesis, and clinical manifestations of celiac disease in adults", section on 'Prognosis'.)
PHENOTYPES — For many years, celiac disease was defined by a set of classic clinical manifestations. However, the combination of serologic, genetic, and histologic data has led to an appreciation of the highly variable clinical manifestations of the condition and the description of other subcategories of celiac disease. Celiac disease can be categorized as classic, non-gastrointestinal, silent/subclinical, or potential, depending on the clinical phenotype (table 5). This classification scheme helps to identify patients who present with nonclassic clinical features and to monitor those who do not meet full diagnostic criteria for celiac disease.
Classic celiac disease — The classic description of celiac disease, or gluten-sensitive enteropathy, includes the following three features:
●Signs of malabsorption such as steatorrhea, weight loss, or other signs of nutrient or vitamin deficiency [135]
●The presence of characteristic histologic changes (including villous atrophy) on small intestinal biopsy
●Resolution of the mucosal lesions and symptoms upon withdrawal of gluten-containing foods, usually within a few weeks to months
Interpretation of the intestinal biopsy is discussed separately. The degree of the villous atrophy does not necessarily correlate with the severity of clinical symptoms. (See "Diagnosis of celiac disease in children", section on 'Subsequent evaluation'.)
Failure to improve on a gluten-free diet is usually due to poor dietary compliance or other underlying malabsorptive disorders. However, in rare cases, diet-refractory celiac disease may be related to sprue-associated lymphoma or to collagenous sprue, a related but little-understood disorder. (See "Management of celiac disease in adults", section on 'Refractory sprue'.)
Non-gastrointestinal celiac disease — In some patients with celiac disease, non-gastrointestinal (also known as extraintestinal) manifestations are predominant (table 3A), with few or no gastrointestinal symptoms. In the past, this was termed "atypical" celiac disease because it was thought to be uncommon, but the advent of serologic screening for celiac disease has identified a substantial proportion of patients with this symptom profile. As for patients with classical disease, the diagnosis requires serologic testing, biopsy evidence of villous atrophy, and improvement of symptoms on a gluten-free diet (table 5).
Silent/subclinical celiac disease — These patients have no discernible symptoms of celiac disease but have a positive specific serologic test for celiac disease and biopsy evidence of villous atrophy. These cases are usually detected by screening of high-risk groups. The term "silent" may be a misnomer; after treatment with a gluten-free diet, many of these patients retrospectively recognize subclinical symptoms that they had not previously considered to be abnormal. (See 'Subclinical disease' above.)
Potential celiac disease — Patients who have positive results to celiac-specific antibodies (eg, antibodies to tissue transglutaminase [anti-tTG]) but have never had a biopsy consistent with celiac disease are said to have "potential" celiac disease (table 5) [136]. Such patients usually are identified during routine testing performed because they are members of a group at increased risk for celiac disease (eg, first-degree relatives of individuals with celiac disease). (See 'High-risk groups' above.)
Observational studies have reported that the majority of asymptomatic children with potential celiac disease will not develop celiac disease during childhood [136,137].
Latent celiac disease — Latent celiac disease describes individuals who have met all of the diagnostic criteria for celiac disease at some time but subsequently recover and have documented normal duodenal/jejunal mucosa and few or no symptoms while on a gluten-containing diet [19,138]. In many such cases, the celiac disease was initially diagnosed during childhood and was treated with a gluten-free diet; then, the disease remained quiescent despite resumption of a normal diet.
The term may also be used to refer to the reverse pattern, in which a patient had documented normal mucosa while ingesting a gluten-containing diet but subsequently developed celiac disease.
Non-celiac gluten sensitivity — Non-celiac gluten sensitivity (NCGS) describes a syndrome of symptomatic response to gluten ingestion in patients with no serologic or histologic evidence of celiac disease. The most common complaints are abdominal pain, bloating, and/or change in bowel patterns, but some patients complain of extraintestinal symptoms. The onset is typically within hours or a few days of ingesting gluten. This time course distinguishes NCGS from the rapid onset of symptoms in wheat allergy (minutes to hours) but can overlap with the delayed onset of symptoms in celiac disease (days to weeks) (table 6) [139].
●Pathophysiology – In many patients with symptoms that they attribute to gluten, gluten is probably not the specific trigger. The symptoms are not replicated on double-blind food challenge, suggesting a placebo or non-gluten-related physiologic effect. An example of the latter are adults whose gastrointestinal symptoms appear to be induced by the fermentable, poorly absorbed short-chain carbohydrates (fermentable oligo-, di-, and monosaccharides and polyols ["FODMAP"]). Since these sugars are also reduced in a gluten-free diet, the clinical response may result from reduction in the oligosaccharides rather than from elimination of the gluten itself [140]. These patients are probably more accurately categorized as having irritable bowel syndrome triggered by oligosaccharides rather than NCGS. (See "Treatment of irritable bowel syndrome in adults", section on 'Gluten avoidance'.)
In a smaller number of patients with these symptoms, it appears that gluten itself is the specific trigger for symptoms, in which case, they are appropriately categorized as NCGS.
The most compelling evidence comes from a double-blind, placebo-controlled crossover food challenge in Italian children with suspected NCGS, based on reported symptomatic response to dietary gluten and lack of serologic evidence of celiac disease or IgE-mediated allergy to wheat or gluten. NCGS was confirmed in 39 percent of subjects, based on the development of symptoms in response to the gluten challenge (with no change in FODMAP content) [141]. NCGS was disproven in the remaining 61 percent. The estimated prevalence of NCGS in children with functional gastrointestinal symptoms was between 0.3 and 1 percent.
Together, these studies suggest that a clinical response to a gluten-free diet may be caused by a variety of mechanisms, including placebo effect and FODMAP reduction, as well as by true NCGS in less than one-half of patients.
●Diagnosis – For children with symptoms that they attribute to gluten, it is important to test for both celiac disease and IgE-mediated wheat allergy. Testing should be performed before eliminating gluten from the diet because these tests may be falsely negative if performed while on a gluten-free diet. The diagnostic evaluation for celiac disease is discussed separately (see "Diagnosis of celiac disease in children", section on 'Initial serologic testing'), including strategies for evaluating patients who are already on a gluten-free diet. (See "Diagnosis of celiac disease in children", section on 'Special populations'.)
The possibility of NCGS may be entertained if celiac disease and wheat allergy have been excluded. However, no tests can reliably distinguish those with true NCGS from those with irritable bowel syndrome or other symptoms that are not specifically related to gluten. Therefore, the diagnosis of NCGS should be approached with caution and should not be based solely on a short-term improvement in symptoms on a gluten-free diet. Neither anti-gliadin antibodies nor other biomarkers can reliably identify patients with NCGS [142]. The diagnosis of NCGS relies on patient-reported symptoms but can be made more objectively by use of a standardized questionnaire before and six weeks after instituting a gluten-free diet, or by use of a double-blind gluten challenge and crossover in patients already on a gluten-free diet [143]. However, a rigorous double-blind gluten challenge is difficult to implement in a clinical setting and is rarely performed except in research studies.
Before a patient with suspected NCGS embarks on a gluten elimination diet, the patient and family should be educated on the possible reasons for a clinical response to this diet and counseled on the potential nutritional deficiencies associated with a gluten-free diet. Involving a knowledgeable dietitian in the care of children who are placed on a gluten-free diet is highly recommended.
SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Celiac disease" and "Society guideline links: Dermatitis herpetiformis".)
INFORMATION FOR PATIENTS — UpToDate offers two types of patient education materials, "The Basics" and "Beyond the Basics." The Basics patient education pieces are written in plain language, at the 5th to 6th grade reading level, and they answer the four or five key questions a patient might have about a given condition. These articles are best for patients who want a general overview and who prefer short, easy-to-read materials. Beyond the Basics patient education pieces are longer, more sophisticated, and more detailed. These articles are written at the 10th to 12th grade reading level and are best for patients who want in-depth information and are comfortable with some medical jargon.
Here are the patient education articles that are relevant to this topic. We encourage you to print or e-mail these topics to your patients. (You can also locate patient education articles on a variety of subjects by searching on "patient info" and the keyword(s) of interest.)
●Basics topics (see "Patient education: Celiac disease (The Basics)")
●Beyond the Basics topics (see "Patient education: Celiac disease in children (Beyond the Basics)")
SUMMARY AND RECOMMENDATIONS
●Pathogenesis – Celiac disease is an immune-mediated inflammatory disease of the small intestine caused by sensitivity to dietary gluten and related proteins in genetically sensitive individuals. The disorder is common, occurring in 0.5 to 1 percent of the general population in most countries. The grains that contain the triggering proteins are wheat, barley, and rye. The small intestinal mucosa improves morphologically when treated with a gluten-free diet and relapses when gluten is reintroduced. (See 'Pathogenesis and risk factors' above.)
●Epidemiology and risk factors – The prevalence of celiac disease is substantially increased in first- and second-degree relatives of patients with celiac disease and in individuals with Down syndrome, type 1 diabetes mellitus, selective IgA deficiency, autoimmune thyroid disease, and probably Turner and Williams syndromes (table 2). Because of the increased risk, we suggest routine screening for celiac disease in pediatric patients in these risk groups, although this approach is somewhat controversial. (See 'High-risk groups' above and "Diagnosis of celiac disease in children", section on 'Indications for testing'.)
●Clinical manifestations
•Gastrointestinal – Classic gastrointestinal symptoms of patients with celiac disease include symptoms of malabsorption such as diarrhea, steatorrhea, weight loss, or other signs of nutrient or vitamin deficiency. (See '"Classical" gastrointestinal symptoms' above.)
•Extra-intestinal – Some patients with celiac disease have non-gastrointestinal manifestations, in addition to or instead of gastrointestinal symptoms. The most specific extraintestinal manifestation is dermatitis herpetiformis, which is uncommon prior to puberty (picture 2A-B). Other extraintestinal manifestations include delayed growth and pubertal development, neurologic disease and behavioral symptoms, arthritis, dental enamel defects, liver disease, and iron deficiency (table 3A-B). (See 'Non-gastrointestinal manifestations' above.)
●Phenotypes of celiac disease
•Classic celiac disease is characterized by gastrointestinal and/or other symptoms outlined above, with associated villous atrophy on small intestinal biopsy (picture 1A-B), and resolution of the clinical symptoms and mucosal lesions upon withdrawal of gluten-containing foods. (See 'Classic celiac disease' above.)
•Other phenotypes of celiac disease are summarized in the table (table 5) (See 'Silent/subclinical celiac disease' above and 'Potential celiac disease' above.)
●Non-celiac gluten sensitivity – This describes a syndrome of symptomatic response to gluten ingestion in patients with no serologic or histologic evidence of celiac disease or wheat allergy. In many patients with symptoms that they attribute to gluten, gluten is probably not a specific trigger. In a smaller number of patients with these symptoms, it appears that gluten itself is the specific trigger for symptoms, in which case, they are appropriately categorized as NCGS. (See 'Non-celiac gluten sensitivity' above.)
