Celiac disease: a known trigger, a measurable antibody, and a test that means different things to different people

Topic: celiac disease: an autoimmune enteropathy with a known trigger, a serological titer that scales with enteropathy, and a diet that reverses it · Since 1950 · Grounded citations only · Published 2026-08-30

Celiac disease is an immune reaction to gluten — the storage protein of wheat, and its relatives in rye and barley — in genetically susceptible people. The reaction damages the lining of the small intestine, flattening the villi that absorb nutrients, and the consequences run from malabsorption and anaemia to bone disease, and to a long list of effects outside the gut.

It is not an allergy, and it is not an intolerance. Wheat allergy is a distinct, IgE-mediated reaction to ingested wheat [1], and the consensus nomenclature separates the three routes explicitly: allergic (wheat allergy), autoimmune (celiac disease, dermatitis herpetiformis, gluten ataxia) and possibly immune-mediated (gluten sensitivity) [2]. Celiac disease is an autoimmune disease: the immune system, having been shown a food protein, attacks one of the body's own enzymes. Gluten is the trigger, but the tissue being destroyed is the patient's own — which is why the damage continues as long as the trigger is present, and why the condition carries the same associations as other autoimmune diseases.

It is common. A nationally representative US survey of 7,798 people put prevalence at 0.71 percent (1 in 141), rising to 1.01 percent among non-Hispanic whites — and of the 35 people found, 29 were unaware of the diagnosis [3]. A large earlier US screening study of 13,145 subjects found 1 in 133 in not-at-risk individuals [4]. Prevalence has been rising over time [5], and the World Gastroenterology Organisation puts adult prevalence between roughly 1 in 100 and 1 in 300 in most of the world [6].

Most cases are undiagnosed, and the reason is that the classical picture — a wasted child with diarrhoea and a distended abdomen — is the minority presentation. Many patients have few symptoms or atypical ones [6], and the terminology has had to be standardised twice to keep up [7] [2]. Screening-detected celiac disease is, on closer questioning, usually not silent — simply unrecognised [8].

The treatment is a strict lifelong gluten-free diet, and it works. That combination — a known trigger, an effective treatment, and most patients undiagnosed — is what makes the diagnostic question in this review the central one.

Start here: why gluten, and why these people

The peptide that cannot be digested

Gluten is unusual as a food protein: glutamine is its most abundant amino acid and proline its second [9], and because gluten proteins are proline-rich they resist enzymatic digestion in the gastrointestinal tract — a property that probably contributes to their immunogenicity [10]. In 2002 a specific 33-amino-acid peptide from α-gliadin was identified with a remarkable set of properties: it is stable against all gastric, pancreatic and intestinal brush-border proteases; it reacts with tissue transglutaminase far more selectively than that enzyme's known natural substrates; it induced gut-derived T-cell lines from 14 of 14 celiac patients tested; and homologues of it are present in every food grain toxic to celiac patients and absent from every non-toxic one [11].

That last clause is the part worth pausing on. The peptide's distribution across grains matches the clinical toxicity of those grains exactly. It is about as close to identifying the molecular culprit as a disease of this kind gets.

Transglutaminase, and why the antibody exists

Tissue transglutaminase was identified as the autoantigen of celiac disease in 1997 [12]. Transglutaminases are cross-linking enzymes with a wide biological role [13] [14]; in the gut, tissue transglutaminase generates T-cell-stimulatory gluten peptides by deamidating glutamine — and only particular glutamine residues are modified, with the spacing between glutamine and proline determining which [9]. That selective deamidation strongly enhances gliadin-specific T-cell reactivity [15].

That modification is what completes the mechanism. The deamidated peptides bind far better to the HLA-DQ2 and HLA-DQ8 molecules on antigen-presenting cells, which display them to CD4 T cells [16] [17]; the adult intestinal T-cell response to α-gliadin focuses on a single deamidated glutamine targeted by transglutaminase [18]. Gliadin-specific T cells restricted by DQ2 were isolated from the small-intestinal mucosa of patients [19], the binding rules for gliadin epitopes on DQ2 have been refined [20], and the equivalent DQ8-restricted peptide was characterised [21] [22]. The autoantibody against transglutaminase is a by-product of that T-cell response — which is precisely why measuring it works as a test.

Genetics

HLA-DQ2 and DQ8 are necessary but not sufficient [23]. They are found in virtually all patients — close to 100 percent sensitivity — and also in 30 to 40 percent of the European population, the great majority of whom will never develop celiac disease [6]. That asymmetry is exactly why HLA typing rules the diagnosis out well and rules it in barely at all. Genome-wide association studies have added many further risk variants influencing immune gene expression [24] [25], with substantial overlap with type 1 diabetes [26] — shared and distinct variants across the two conditions, which is the genetic basis for their clinical co-occurrence. Environmental triggers have been investigated, including early-childhood rotavirus infection [27].

Pillar 1: measurement and diagnosis

Serology first

The first-line test is IgA antibody against tissue transglutaminase (tTG-IgA), and in a paediatric comparison against histology it was the best available single index — sensitivity 92.5 percent, specificity 97.6 percent [28]. Antibodies against deamidated gliadin peptides and endomysial antibodies are the other members of the panel [29] [30] [31].

Two practical traps matter more than the assay details.

The first is IgA deficiency. tTG-IgA cannot be positive in someone who makes no IgA. In one audit of 9,533 patients tested for endomysial antibodies, only 4,698 (49 percent) were also tested for IgA deficiency; deficiency was found in 35 of those 4,698 (0.75 percent), and only 19 of the 35 (54 percent) were then managed appropriately [32]. This is why a total IgA is ordered alongside the tTG, and why an IgG-based test is used when IgA is absent — in one study IgG anti-gliadin correctly identified all three children with celiac disease and total IgA deficiency whose IgA-based tests were negative [28]. Whether total IgA must be measured in everyone has been debated [33].

The second trap is the one patients most often fall into: the tests only work while you are eating gluten. A gluten-free diet normalises the antibodies and heals the mucosa, so someone who removes gluten before being tested may become untestable for months. This is not a technicality — antibody titres in treated patients fall below the cutoff over the following year [28], and a formal gluten-challenge study showed measurable histological damage returning within 14 days of resuming gluten, with antibody titres rising markedly by day 28 [34]. The practical rule is to test first and change the diet afterwards.

Biopsy, and when it can be skipped

Confirmation is by duodenal biopsy showing the characteristic changes — increased intraepithelial lymphocytes, crypt hyperplasia and villous atrophy, graded on the Marsh classification [35]. The European diagnostic criteria were formalised in 1990 [36] and revised as the antibody tests improved.

Two caveats about the biopsy itself. Villous atrophy in adults can be patchy, so multiple samples are needed and a single negative biopsy is not conclusive [37]; and lesser degrees of change — raised intraepithelial lymphocytes at the villous tip without atrophy — are hard to interpret [38].

The no-biopsy pathway. The 2011 ESPGHAN guidelines introduced an option that has changed paediatric practice: in a child with symptoms suggestive of celiac disease and tTG-IgA above ten times the upper limit of normal, the diagnosis may be made without duodenal biopsy, under a strict protocol with further laboratory tests including endomysial antibodies and HLA typing [39]. Later European guidance retains this for a defined group of children [40]. The logic is the subject of the figure below, and it is entirely about predictive value.

HLA typing is useful mainly in the negative direction: celiac disease is unlikely in someone with neither DQ2 nor DQ8 [39]. It is a good rule-out test and a poor rule-in one, for the same reason the antibody is a good rule-out test in low-prevalence settings.

What celiac disease is not

Wheat allergy is a distinct IgE-mediated condition, and the amounts of gluten that are safe differ between the two [41]. Non-celiac gluten sensitivity is the harder category. A double-blind, randomised, placebo-controlled rechallenge trial in patients with irritable bowel syndrome who had celiac disease excluded and were symptomatically controlled on a gluten-free diet found that gluten did induce symptoms: of 19 patients rechallenged with gluten, 13 (68 percent) reported that symptoms were not adequately controlled [42]. The mucosal immunology differs from celiac disease — IL-17 expression separates the two gliadin-induced disorders [43] — and the nomenclature has been formalised [2] [44]. Gluten-related neurological syndromes, including cerebellar ataxia, have their own literature and their own serological marker in transglutaminase 6 antibodies [45] [46] [47].

Centerpiece: what a positive antibody actually means

Sensitivity and specificity are properties of an assay. Predictive value is a property of the assay and the population together, and celiac disease is an unusually clean illustration of why the difference matters.

The inputs are all measured, in separate studies. tTG-IgA: sensitivity 92.5 percent, specificity 97.6 percent, in 161 children with biopsy-confirmed disease and 129 controls [28]. Prevalence: 1 in 22 in first-degree relatives, 1 in 39 in second-degree relatives, 1 in 56 in symptomatic patients and 1 in 133 in not-at-risk individuals, from 13,145 people screened [4]. The only operation applied to them is Bayes' theorem.

0.2 0.5 1 2 5 10 20 50 pre-test probability of celiac disease (%, log scale) 0 20 40 60 80 100 positive predictive value (%) 90% — still one false positive in ten general population at 1%   2 8 % :   s e v e n   i n   t e n positives would be false v a l i d a t i o n   s t u d y ' s   o w n   m i x     9 8 % One assay, sensitivity 92.5% and specificity 97.6%. What a positive means depends on who was tested. 0 20 40 60 80 100 positive predictive value (%) validation study (161 vs 129) 1st-degree relative 1 in 22 2nd-degree relative 1 in 39 symptomatic 1 in 56 not at risk 1 in 133 general population at 1% Prevalences: 1 in 22, 1 in 39, 1 in 56 and 1 in 133 are measured values from 13,145 people screened in the United States. The 1% figure is the one the World Gastroenterology Organisation names when it says tTG has a low positive predictive value in the general population — an adjective this arithmetic turns into 28%. Prevalence, not the assay, decides 98% 65% 50% 41% 23% 28%
Left: positive predictive value of tTG-IgA against pre-test probability, on a log scale, with the measured prevalences marked. Right: the same assay across real populations.

The check the arithmetic was never given. That validation paper reports a sensitivity, a specificity, a positive predictive value of 98 percent and a negative predictive value of 91.2 percent — and separately reports that it enrolled 161 cases and 129 controls. It never connects them. Applying Bayes at the study's own case-control ratio of 55.5 percent gives a PPV of 97.96 percent and an NPV of 91.25 percent, reproducing both headline figures to within a tenth of a percentage point. That is not a coincidence and it is not a compliment: it demonstrates that the reported predictive values are a property of the enrolment design — roughly half cases by construction — and therefore do not transfer to any clinic.

A second check, against a different source. The World Gastroenterology Organisation states that tTG has a "low positive predictive value in the general (low risk) population (with a prevalence of 1%)", and gives no number [6]. Using sensitivity and specificity measured by an unrelated group, Bayes puts it at 28 percent — so roughly seven in ten positives in an unselected population would be false. The guideline's adjective is confirmed by arithmetic that neither paper performed.

What follows from that. The same assay, unchanged, gives a positive predictive value of 98 percent in its validation study, 65 percent in a first-degree relative, 41 percent in a symptomatic patient, 23 percent in someone not at risk, and 28 percent in an unselected population at 1 percent prevalence. Nothing about the test changed. Only who was tested.

This is the whole justification for three things that otherwise look like bureaucracy. It is why biopsy confirmation is standard — even a 65 percent PPV in the highest-risk group is not a diagnosis. It is why guidelines recommend a serial algorithm adding the more specific endomysial antibody in low-prevalence settings [6]. And it is why the no-biopsy pathway requires a very high titre: raising the threshold to ten times the upper limit of normal trades sensitivity for specificity, and specificity is exactly the term that governs predictive value when prevalence is low [39].

The mirror image is the reason these tests are still excellent. At 1 percent prevalence the negative predictive value exceeds 99.9 percent. tTG-IgA is close to conclusive for ruling celiac disease out and weak for ruling it in — which is the correct shape for a first-line test in a common, under-diagnosed, easily-missed condition.

Three honest limits. The sensitivity and specificity come from a paediatric cohort and would differ in adults and across assay manufacturers. The figure assumes a single fixed cut-off, whereas the real diagnostic pathway uses the titre as a continuous variable — which is the whole point of the 10× rule, and this substrate does not contain titre-stratified predictive values, so the figure cannot show how the curve shifts at higher thresholds. And pre-test probability in practice is a clinical judgement, not one of the four tidy numbers plotted here.

Pillar 2: treatment

The gluten-free diet

There is one treatment and it is dietary: strict, lifelong exclusion of the gluten-containing cereals — wheat, rye and barley [6] [48]. It heals the intestine, resolves symptoms in most patients, and normalises the antibodies.

How completely it heals, and how fast, is worth knowing precisely. In 158 patients followed with serial biopsies over 15 years, histological remission among those with villous atrophy was 65.0 percent within 2 years, 85.3 percent within 5 years, and 89.9 percent in long-term follow-up — leaving 10.1 percent with persisting villous atrophy after 5 years. Children did far better: 95 percent recovered within 2 years and 100 percent in the long term [49]. Adults heal slowly and sometimes incompletely, and the gap between symptom resolution and mucosal healing is real.

Adherence is the limiting factor. A systematic review of 38 studies found strict adherence ranging from 42 to 91 percent depending on definition and measurement, lowest among ethnic minorities and among those diagnosed in childhood, and most strongly associated with cognitive, emotional and socio-cultural factors, membership of an advocacy group, and regular dietetic follow-up [50]. Notably, patients found by screening adhered no worse than those who presented with symptoms — an argument against the assumption that asymptomatic people will not comply. Adherence is also expensive: the economic burden of the diet itself has been quantified [51]. Objective measurement is now possible through gluten immunogenic peptides detectable in urine [52].

Nutrition and bone

Malabsorption leaves deficits that need looking for — and they persist. Poor vitamin status was documented in celiac patients who had been on a gluten-free diet for ten years [53], so the diet does not automatically fix nutrition.

Bone deserves specific attention. Bone mass and metabolism are abnormal in celiac disease [54] [55], the gluten-free diet improves bone mineral content in growing patients [56], and there are joint guidelines for osteoporosis in celiac disease and inflammatory bowel disease [57]. Conversely, celiac disease is found more often than expected among patients presenting with osteoporosis, prompting arguments for routine screening in that group [58].

Monitoring, and refractory disease

Follow-up tracks antibody normalisation and, where indicated, mucosal healing. Clinical and laboratory information have a high positive but low negative predictive value for persisting intestinal damage in treated patients, and dietary compliance assessed by interview was the best single marker of control [59] — another instance of the theme running through this review.

Refractory celiac disease — persisting villous atrophy and malabsorption despite a strict diet — affected 7.0 percent of that follow-up cohort, and 5 of those 11 patients developed enteropathy-associated T-cell lymphoma [49]. It is classified by whether the intraepithelial lymphocytes are phenotypically abnormal [60] [61] [62], the relationship to lymphoma is well described [63], and immunosuppression with azathioprine and prednisone has been used [64]. Lymphoproliferative malignancy risk tracks the small-intestinal histopathology [65], as does mortality [66] [67] [68] [69].

Associated conditions and relatives

Celiac disease travels with other autoimmune disease — autoimmune thyroid disease [70] and type 1 diabetes above all, with celiac disease appearing after the onset of type 1 diabetes in prospective follow-up [71] and the two sharing genetic variants [26]. Duration of gluten exposure has been examined as a risk factor for developing further autoimmune disorders [72]. Severe liver disease is another association in which the diet may reverse the abnormality [73].

Testing first-degree relatives is justified by the arithmetic above: at 1 in 22, they are the group in which a positive result carries the most weight [4].

Drugs

There are none in routine use. The gluten-free diet remains the only treatment [40] [74] [48].

Pillar 3: what is unresolved

Non-dietary therapy. Three strategies have reached patients in trials. Enzyme therapy follows directly from the 33-mer's protease resistance — the original paper showed the peptide could be detoxified by a bacterial prolyl endopeptidase and proposed oral peptidase supplementation on that basis [11], combination enzyme therapy for gastric digestion of dietary gluten has been tested in patients [75], and a prolyl endoprotease has been shown to degrade gluten efficiently [10]. Tight-junction modulation with larazotide acetate was studied for preventing celiac activation during gluten challenge [76] and for persistent symptoms despite a gluten-free diet [77] [78]. None is standard care, and the honest framing is that these are adjuncts to the diet rather than replacements for it.

Mass screening. The case is unusually strong on paper — the disease is common, most cases are undiagnosed [3], screening-detected patients are usually symptomatic once asked [8], and they adhere to the diet as well as anyone [50]. Screening has been done sequentially with serology and gastroscopy in North America [79], in primary care with rapid tests [80], and cost-effectiveness has been modelled in defined high-risk groups such as children with Down syndrome [81]. What the figure above adds is the counterargument: at population prevalence, most positives are false, so any mass programme must be a two-stage algorithm and must be costed as one.

Better diagnostics. Deamidated gliadin peptide and tTG antibodies have been meta-analysed against each other [29], urinary gluten immunogenic peptides offer objective adherence measurement [52], and the open question is a titre-based pathway validated in adults as the 10× rule has been in children [39] [40].

Who heals and who does not. One in ten adults still has villous atrophy after five years of a strict diet [49], and predicting them in advance is not currently possible.

Dig deeper in lmmol

Celiac disease belongs with the autoimmune cluster and the associations are clinical, not merely thematic. Thyroid disease is the commonest companion [70], and type 2 diabetes and glycemic control is the nearest published relative of the type 1 diabetes link, where the two conditions share genetic variants outright [26] [71]. Inflammatory bowel disease is the instructive contrast: another immune-mediated enteropathy, but one with no identified trigger — remove gluten and celiac disease resolves, whereas Crohn's and colitis are managed by suppressing immunity indefinitely, and the two conditions even share osteoporosis guidance [57]. Rheumatoid arthritis and psoriasis round out the autoimmune siblings, and celiac disease is the one among them where the antigen is known and avoidable. Osteoporosis is a downstream consequence worth reading in both directions, since undiagnosed celiac disease turns up among people presenting with low bone density [58] [57]. The full collection is at health.

Key papers

  1. W4246138613: Wheat allergy (cited 145×)
  2. W2118019845: Spectrum of gluten-related disorders: consensus on new nomenclature and classification (cited 1,148×)
  3. W2090331375: The Prevalence of Celiac Disease in the United States (cited 748×)
  4. W2119425622: Prevalence of celiac disease in at-risk and not-at-risk groups in the United States: a large multicenter study. (cited 1,706×)
  5. W1544004437: Increasing prevalence of coeliac disease over time (cited 751×)
  6. W2030602908: World Gastroenterology Organisation Global Guidelines on Celiac Disease (cited 268×)
  7. W2116287226: The Oslo definitions for coeliac disease and related terms (cited 1,715×)
  8. W2100294153: Coeliac disease detected by screening is not silent--simply unrecognized (cited 129×)
  9. W2119520032: Specificity of Tissue Transglutaminase Explains Cereal Toxicity in Celiac Disease (cited 360×)
  10. W2096941822: Efficient degradation of gluten by a prolyl endoprotease in a gastrointestinal model: implications for coeliac disease (cited 260×)
  11. W1531267945: Structural Basis for Gluten Intolerance in Celiac Sprue (cited 1,587×)
  12. W2043856196: Identification of tissue transglutaminase as the autoantigen of celiac disease (cited 2,065×)
  13. W2067403820: Transglutaminases: Nature’s biological glues (cited 1,047×)
  14. W4211195301: Transglutaminases (cited 594×)
  15. W1641758654: Selective deamidation by tissue transglutaminase strongly enhances gliadin-specific T cell reactivity. (cited 493×)
  16. W1576698873: Gluten, major histocompatibility complex, and the small intestine. A molecular and immunobiologic approach to the spectrum of gluten sensitivity ('celiac sprue'). (cited 2,011×)
  17. W4294305087: Gluten, major histocompatibility complex, and the small intestine (cited 1,707×)
  18. W2113819336: The Intestinal T Cell Response to α-Gliadin in Adult Celiac Disease Is Focused on a Single Deamidated Glutamine Targeted by Tissue Transglutaminase (cited 661×)
  19. W2036052530: Gliadin-specific, HLA-DQ(alpha 1*0501,beta 1*0201) restricted T cells isolated from the small intestinal mucosa of celiac disease patients. (cited 599×)
  20. W1954115773: Refining the Rules of Gliadin T Cell Epitope Binding to the Disease-Associated DQ2 Molecule in Celiac Disease: Importance of Proline Spacing and Glutamine Deamidation (cited 169×)
  21. W2124247925: An immunodominant DQ8 restricted gliadin peptide activates small intestinal immune response in in vitro cultured mucosa from HLA-DQ8 positive but not HLA-DQ8 negative coeliac patients (cited 91×)
  22. W2063044744: T cells from the small intestinal Mucosa of a DR4, DQ7/DR4. DQ8 celiac disease patient preferentially recognize gliadin when presented by DQ8 (cited 187×)
  23. W2048858087: Genetic Background of Celiac Disease and Its Clinical Implications (cited 208×)
  24. W1991443138: A genome-wide association study for celiac disease identifies risk variants in the region harboring IL2 and IL21 (cited 675×)
  25. W2112409097: Multiple common variants for celiac disease influencing immune gene expression (cited 1,065×)
  26. W2061446718: Shared and Distinct Genetic Variants in Type 1 Diabetes and Celiac Disease (cited 759×)
  27. W1966788203: Rotavirus Infection Frequency and Risk of Celiac Disease Autoimmunity in Early Childhood: A Longitudinal Study (cited 562×)
  28. W1997917153: Antibodies against Synthetic Deamidated Gliadin Peptides for Celiac Disease Diagnosis and Follow-Up in Children (cited 84×)
  29. W2028455717: Meta‐analysis: deamidated gliadin peptide antibody and tissue transglutaminase antibody compared as screening tests for coeliac disease (cited 221×)
  30. W2160550562: Tissue transglutaminase autoantibody enzyme-linked immunosorbent assay in detecting celiac disease (cited 611×)
  31. W1968014632: Update on Serologic Testing in Celiac Disease (cited 359×)
  32. W2130516189: Celiac Disease and IgA Deficiency: Complications of Serological Testing Approaches Encountered in the Clinic (cited 111×)
  33. W1996563985: Do we need to measure total serum IgA to exclude IgA deficiency in coeliac disease? (cited 34×)
  34. W2138178372: Kinetics of the histological, serological and symptomatic responses to gluten challenge in adults with coeliac disease (cited 239×)
  35. W2036798664: The histopathology of coeliac disease (cited 1,707×)
  36. W4245464873: Revised criteria for diagnosis of coeliac disease. Report of Working Group of European Society of Paediatric Gastroenterology and Nutrition. (cited 1,329×)
  37. W2069170359: Patchy villous atrophy in adult patients with suspected gluten-sensitive enteropathy: is a multiple duodenal biopsy strategy appropriate? (cited 166×)
  38. W2088154498: Intraepithelial lymphocytes in the villous tip: do they indicate potential coeliac disease? (cited 122×)
  39. W2100548379: European Society for Pediatric Gastroenterology, Hepatology, and Nutrition Guidelines for the Diagnosis of Coeliac Disease (cited 2,699×)
  40. W2937921992: European Society for the Study of Coeliac Disease (ESsCD) guideline for coeliac disease and other gluten‐related disorders (cited 1,010×)
  41. W2169587065: Review article: safe amounts of gluten for patients with wheat allergy or coeliac disease (cited 255×)
  42. W1972526677: Gluten Causes Gastrointestinal Symptoms in Subjects Without Celiac Disease: A Double-Blind Randomized Placebo-Controlled Trial (cited 713×)
  43. W2133246360: Differential Mucosal IL-17 Expression in Two Gliadin-Induced Disorders: Gluten Sensitivity and the Autoimmune Enteropathy Celiac Disease (cited 245×)
  44. W1592644639: Coeliac disease and gluten sensitivity (cited 174×)
  45. W2164941466: Sporadic cerebellar ataxia associated with gluten sensitivity (cited 172×)
  46. W1992303087: Gluten sensitivity: from gut to brain (cited 400×)
  47. W2084359554: Transglutaminase 6 antibodies in the diagnosis of gluten ataxia (cited 159×)
  48. W2962821767: Celiac disease: a comprehensive current review (cited 1,106×)
  49. W2107957285: Histologic Follow-up of People With Celiac Disease on a Gluten-Free Diet (cited 337×)
  50. W1975094828: Systematic review: adherence to a gluten‐free diet in adult patients with coeliac disease (cited 449×)
  51. W2139114869: Economic burden of a gluten‐free diet (cited 316×)
  52. W2203036197: Detection of gluten immunogenic peptides in the urine of patients with coeliac disease reveals transgressions in the gluten-free diet and incomplete mucosal healing (cited 324×)
  53. W2090456989: Evidence of poor vitamin status in coeliac patients on a gluten‐free diet for 10 years (cited 326×)
  54. W2117517579: Bone mass and metabolism in patients with celiac disease (cited 237×)
  55. W2008717439: Influence of pattern of clinical presentation and of gluten-free diet on bone mass and metabolism in adult coeliac disease (cited 132×)
  56. W1859602657: Effect of gluten-free diet on bone mineral content in growing patients with celiac disease (cited 144×)
  57. W2002592715: Guidelines for osteoporosis in coeliac disease and inflammatory bowel disease (cited 383×)
  58. W2110696339: Increased Prevalence of Celiac Disease and Need for Routine Screening Among Patients With Osteoporosis (cited 182×)
  59. W1976374163: Long-Term Follow-Up of Celiac Adults on Gluten-Free Diet: Prevalence and Correlates of Intestinal Damage (cited 250×)
  60. W2149711457: Abnormal intestinal intraepithelial lymphocytes in refractory sprue (cited 404×)
  61. W2088286900: Distinction between coeliac disease and refractory sprue: a simple immunohistochemical method (cited 165×)
  62. W2156197620: Classification and management of refractory coeliac disease (cited 333×)
  63. W2109792437: Refractory sprue, coeliac disease, and enteropathy-associated T-cell lymphoma (cited 759×)
  64. W2008322346: Azathioprine and prednisone combination therapy in refractory coeliac disease (cited 193×)
  65. W2107832827: Risk of Lymphoproliferative Malignancy in Relation to Small Intestinal Histopathology Among Patients With Celiac Disease (cited 154×)
  66. W1997027964: Small-Intestinal Histopathology and Mortality Risk in Celiac Disease (cited 363×)
  67. W2097723680: Mortality in patients with coeliac disease and their relatives: a cohort study (cited 600×)
  68. W2128353487: Coeliac disease--associated disorders and survival. (cited 474×)
  69. W1889062537: Mortality in celiac disease (cited 293×)
  70. W2108178123: Coeliac disease and autoimmune thyroid disease. (cited 202×)
  71. W2067309350: Occurrence of Celiac Disease After Onset of Type 1 Diabetes: A 6-Year Prospective Longitudinal Study (cited 275×)
  72. W1987433373: Duration of exposure to gluten and risk for autoimmune disorders in patients with celiac disease (cited 904×)
  73. W2086059406: Celiac disease in patients with severe liver disease: Gluten-free diet may reverse hepatic failure (cited 308×)
  74. W2131017558: Diagnosis and management of adult coeliac disease: guidelines from the British Society of Gastroenterology (cited 1,122×)
  75. W2034141008: Combination Enzyme Therapy for Gastric Digestion of Dietary Gluten in Patients With Celiac Sprue (cited 237×)
  76. W2039229476: A Randomized, Double-Blind Study of Larazotide Acetate to Prevent the Activation of Celiac Disease During Gluten Challenge (cited 190×)
  77. W2162876267: Larazotide Acetate for Persistent Symptoms of Celiac Disease Despite a Gluten-Free Diet: A Randomized Controlled Trial (cited 263×)
  78. W2038805797: Cell polarity-determining proteins Par-3 and PP-1 are involved in epithelial tight junction defects in coeliac disease (cited 132×)
  79. W2000774337: Screening for Celiac Disease in a North American Population: Sequential Serology and Gastrointestinal Symptoms (cited 150×)
  80. W2017600139: Population screening for coeliac disease in primary care by district nurses using a rapid antibody test: diagnostic accuracy and feasibility study (cited 150×)
  81. W2080143666: Screening for Celiac Disease in Asymptomatic Children With Down Syndrome: Cost-effectiveness of Preventing Lymphoma (cited 88×)