SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. EPIDEMIOLOGY
  5. GLUTEN
  6. GENETIC PREDISPOSITION
  7. CLINICAL FEATURES AND DIAGNOSIS
  8. DIAGNOSTIC CRITERIA
  9. SEROLOGICAL MARKERS
  10. TREATMENT
  11. NONRESPONSIVE COELIAC DISEASE
  12. COMPLICATIONS OF CD
  13. FOLLOW-UP
  14. References

Coeliac disease (CD) is a disorder of the small intestine, characterized by villous atrophy, due to an intolerance to dietary gluten in genetically susceptible individuals, which responds to gluten withdrawal. The underlying immunological mechanisms causing the disorder are still being worked out.

In recent years a wide range of clinical presentations has become increasingly apparent, as has a lengthening list of associated conditions. Severe malabsorption with steatorrhoea and profound weight loss is seen infrequently, perhaps as a result of earlier diagnosis and the recognition of ‘silent’ and ‘latent’ disease.

The prevalence of CD as judged by population screening with, in particular, anti-endomysial antibodies, appears to be much higher than that found with clinically apparent cases.

There are a variety of well recognized complications, the commonest probably being osteopenia and osteoporosis. The marked increased risk of lymphoma can be avoided by a strict gluten-free diet.

Follow-up of patients needs to be lifelong with prompt investigation of new symptoms and blood test abnormalities.


INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. EPIDEMIOLOGY
  5. GLUTEN
  6. GENETIC PREDISPOSITION
  7. CLINICAL FEATURES AND DIAGNOSIS
  8. DIAGNOSTIC CRITERIA
  9. SEROLOGICAL MARKERS
  10. TREATMENT
  11. NONRESPONSIVE COELIAC DISEASE
  12. COMPLICATIONS OF CD
  13. FOLLOW-UP
  14. References

Coeliac disease (CD) or gluten-sensitive enteropathy is a chronic malabsorptive disorder of the small intestine, caused by exposure to dietary gluten in genetically susceptible individuals. Histologically it is characterized by villous atrophy, a lowering of the villous height to crypt depth ratio (normal 3–5:1), and an increase in intraepithelial lymphocytes (normal 10–30 per 100 epithelial cells). Clinically there is a wide range of presentation from asymptomatic through fatigue, vague abdominal symptoms, weight loss and diarrhoea to frank malabsorption with steatorrhoea and nutrient deficiences, although the more severe end of the spectrum is increasingly less common. Symptoms and abnormal small bowel histology resolve on removal of gluten from the diet.

EPIDEMIOLOGY

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. EPIDEMIOLOGY
  5. GLUTEN
  6. GENETIC PREDISPOSITION
  7. CLINICAL FEATURES AND DIAGNOSIS
  8. DIAGNOSTIC CRITERIA
  9. SEROLOGICAL MARKERS
  10. TREATMENT
  11. NONRESPONSIVE COELIAC DISEASE
  12. COMPLICATIONS OF CD
  13. FOLLOW-UP
  14. References

The age of presentation and prevalence of CD appears to have changed dramatically over the last 30–40 years.1 It was previously considered to be a disease of childhood, with the majority of diagnoses occurring in children aged less than 2 years. However in Europe in the 1970s and 1980s, there has been an increase in the age of paediatric diagnosis to 4 years, except in Sweden where a dramatic increase in infant cases has been seen.2 The causes of this latter phenomenon are still being debated.3, 4 Adult presentation is increasingly common, and indeed CD can occur at any age.

Quoted clinical case prevalence rates vary widely, being ≈ 1 in 1200 in the UK but of the order of 1 in 300 in western Ireland. However, this has not been reflected in prevalence rates based on population screening studies, which have demonstrated similar rates throughout Europe. In the largest such study, 17 000 school children throughout Italy, aged 6–15, were screened in a stepwise fashion with anti-gliadin antibodies (AGA), anti-endomysial antibodies (AEA), and finally duodenal biopsy in positives.5 A prevalence rate of 1 in 184 was found, with a ratio of 1:7 of known to previously undiagnosed cases. The authors concluded that CD is one of the commonest lifelong disorders in Italy.

Other screening studies have also demonstrated a much higher prevalence than found with diagnosed symptomatic cases. A Swedish study involving blood donors found a prevalence rate of at least 1 in 256.6 A rate of 1 in 250, based on positive AEA in blood donors, was found in a US study,7 although this was not confirmed with histological diagnosis. In Northern Ireland, screening of 1823 adults revealed a prevalence of 1 in 152.8

Recently the concept of the ‘coeliac iceberg’ (Figure 1 ) has helped to reconcile the large differences between identified case prevalence rates and the much higher rates found through population screening.9, 10 This suggests that the majority of people with CD go clinically undetected, having either ‘silent’ or ‘latent’ disease. Silent CD refers to those people with coeliac enteropathy but with no/minor undiagnosed symptoms, and latent CD to people with a normal small bowel on a normal diet, but who have in the past had a flat jejunal biopsy, or will in the future develop such changes, responsive to gluten withdrawal.11 Such subjects may have subtle changes, such as an increased intraepithelial lymphocyte count, but with architecturally normal villi and crypts and whose biopsies may therefore be reported as normal. This latter group is obviously more difficult to demonstrate definitively, and indeed its existence may be disputed, but what is clear is that older concepts of the frequency of gluten sensitivity based on clinical identification of cases are too simplistic and obscure the true nature of the problem.

image

Figure 1. . The CD iceberg and spectrum of gluten sensitivity.

Download figure to PowerPoint

The visible tip of the iceberg is therefore relatively small, but will no doubt increase in size as the wide spectrum of often minor, monosymptomatic and extra-intestinal manifestations of gluten sensitivity become more widely appreciated.

GLUTEN

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. EPIDEMIOLOGY
  5. GLUTEN
  6. GENETIC PREDISPOSITION
  7. CLINICAL FEATURES AND DIAGNOSIS
  8. DIAGNOSTIC CRITERIA
  9. SEROLOGICAL MARKERS
  10. TREATMENT
  11. NONRESPONSIVE COELIAC DISEASE
  12. COMPLICATIONS OF CD
  13. FOLLOW-UP
  14. References

The classical features of CD were first described by Samuel Gee in 1888,12 but it was not until 1950 that its association with certain cereals was recognized by the Dutch paediatrician, Dicke.13 Later he and others narrowed down the nature of the toxic element to the alcohol-soluble component (gliadin) of the gluten fraction of wheat (Figure 2 ).14, 15 It has since been demonstrated that the equivalent prolamin fractions of rye (secalins), barley (hordeins) and probably oats (aveneins) are toxic to CD patients. Whether or not oats are harmful remains controversial. Several recent studies have indicated that a moderate amount of oats (e.g. 50 g/day) is not harmful in adults with CD or dermatitis herpetiformis (DH).16[17]–18 However, the prolamin fraction of oats contains amino acid sequences (–QQQPF–) identical to those present in A-gliadin and which form part of what is felt to be a toxic epitope from that protein.19, 20 Their presence in a much smaller proportion in oats may be the reason for the apparent non-toxicity. This does not discount the possibility, however, that some patients may be exquisitely sensitive to oat prolamin (as can be the case with gliadin), and also that oats can become contaminated with wheat gluten during harvesting and milling. It is the current practice of the authors to advise against the use of oats in the diet of gluten-sensitive patients.21

image

Figure 2. . The subfractions of wheat protein. Gliadins form the alcohol-soluble component of gluten.

Download figure to PowerPoint

GENETIC PREDISPOSITION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. EPIDEMIOLOGY
  5. GLUTEN
  6. GENETIC PREDISPOSITION
  7. CLINICAL FEATURES AND DIAGNOSIS
  8. DIAGNOSTIC CRITERIA
  9. SEROLOGICAL MARKERS
  10. TREATMENT
  11. NONRESPONSIVE COELIAC DISEASE
  12. COMPLICATIONS OF CD
  13. FOLLOW-UP
  14. References

CD is an HLA-associated condition, the primary association being with the MHC class II alleles DQA1*0501 and DQB1*0201.22 This allele combination, which codes for the product HLA DQ2, is found in 98% of CD patients from northern Europe and can be inherited in cis fashion in DR3 homozygotes or trans fashion in DR5, DR7 heterozygotes (i.e. DQA and B alleles encoded on different chromosomes). The association with DR3 is thought to be through linkage disequilibrium with DQ2. In southern Europe DQ2 is also the major susceptibility genotype, being present in 92% of CD patients, but DQ8 also confers susceptibility in DR4 positive individuals.23

Despite this very strong HLA association there is only about 30% concordance for CD in HLA identical siblings, with concordance approaching 100% in monozygotic twins and the risk in first-degree relatives being between 10 and 20%.24, 25 It should also be recognized that ≈ 25% of the ‘normal’ population possess DQ2. Thus other genetic influences, probably non-HLA linked, are very probably required for development of the disease. A genome search for such a gene/genes has been undertaken,26 but as of yet there is no conclusive result.

The known antigen, i.e. gluten, and the HLA predisposition are brought together in the current immune theory of CD pathogenesis. A full discussion of this is beyond the scope of this review but in brief it suggests that gluten-sensitive small intestinal T-cells recognize gluten-derived peptide epitopes when presented in association with DQ2.27 Evidence suggests that activation of CD4+ T-cells in this way leads to a Th1/Th0 type inflammatory response28 and the observed mucosal damage. The exact epitope(s) within gluten is still not known, although there is mounting evidence for the role of a 19 mer peptide from the N-terminal region of A-gliadin (a constituent of gluten). This has been shown to cause the histological features of CD by in vitro organ culture20 and in challenge studies on treated patient volunteers.19 It has also been shown to bind to DQ2 molecules,29, 30 and stimulate T-cells, derived from the peripheral blood of CD patients, when presented in association with DQ2.31

CLINICAL FEATURES AND DIAGNOSIS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. EPIDEMIOLOGY
  5. GLUTEN
  6. GENETIC PREDISPOSITION
  7. CLINICAL FEATURES AND DIAGNOSIS
  8. DIAGNOSTIC CRITERIA
  9. SEROLOGICAL MARKERS
  10. TREATMENT
  11. NONRESPONSIVE COELIAC DISEASE
  12. COMPLICATIONS OF CD
  13. FOLLOW-UP
  14. References

The original description of CD, by Gee,12 was of steatorrhoea and cachexia occurring mainly in children 1–5 years old. As touched on above, the clinical manifestations of CD appear to have changed greatly in the last 40 years or so, with much milder forms being recognized and the extreme end of the spectrum, as noted by Gee, rarely encountered. Whether this is entirely due to greater appreciation of the protean presentations of CD, with cases being diagnosed (especially since the advent of widely available small intestinal biopsy, and anti-endomysial antibodies) that previously would have been overlooked32 or due to a change in the disease itself through some environmental factor(s),1 is debatable. Probably a combination of these reasons is important. The diagnosis of CD remains a histological one, however, and the problem for the clinician is which patients to select for small intestinal biopsy (SIB), when manifestations can be so non-specific and varied.

Table 1 lists some of the clinical features that can be encountered in CD. With adult presentation, weight loss, diarrhoea, lassitude and anaemia will often arouse suspicion but it is when these features are absent that the diagnosis is more difficult. In an Italian study 226 consecutive cases of CD were examined.33 The rate of diagnosis rose dramatically over the course of the 18 years studied (there being only 36 diagnoses in the first half), partly attributed to the finding of subclinical cases—defined in this instance as the absence of diarrhoea and/or weight loss—to the extent that half of all diagnoses in the final 3 years were subclinical and in the last year occurred twice as frequently as ‘classical’ diagnoses. This was in a centre with a special interest, where cases were presumably actively sought, but serves to illustrate the problems of the ‘iceberg’ for the general gastroenterologist, or indeed other physicians, because gluten-sensitive patients may present at a variety of specialist clinics as a result of the increasing number of recognized associated conditions (Table 2 2 ).

Table 1.  . Clinical features encountered in CD Thumbnail image of
Table 2.  . Some of the disorders associated with CD Thumbnail image of

One of the most important of these associated conditions is insulin dependent diabetes mellitus (IDDM), which has been shown to have a strong association with CD in several studies.34, 35 Recently Collin et al.36 studied 335 CD patients and found IDDM in 5.4%, compared to a frequency of 1.5% in an age- and sex-matched control population. This study also found a striking increase in the frequency of Sjogren’s syndrome, being present in 3.3% of the CD patients and in only 0.3% of the controls.

An association with epilepsy has been found in some studies.37, 38 A specific association with epilepsy and cerebral calcification has been reported from Italy,39 but this series used a highly selected group of patients, some of whom had been previously described. Very recently a much larger study from Ireland40 found a frequency of CD of 1 in 44 amongst 177 patients attending a seizure clinic. No association with cerebral calcification was found. The prevalence of CD in a control group of 488 pregnant women was 1 in 244, a figure highly compatible with prevalence rates found in population screening studies (see above).

It is clear that a high index of suspicion needs to be maintained by physicians seeing patients with a variety of conditions, especially as the majority who turn out also to have CD have no gastrointestinal symptoms. Whether all patients with IDDM, Sjogren’s syndrome, etc. should be screened serologically for CD is debatable, but there is a strong case for this course of action. Not only will treatment by way of a gluten-free diet (GFD) frequently improve quality of life, even in apparently asymptomatic patients, there is some evidence that a GFD results in improved control of the associated condition, and importantly will also reduce to normal the greatly increased risk of malignancy associated with untreated CD.41

Dermatitis herpetiformis deserves special mention, because it can be considered as an extraintestinal manifestation of CD rather than an associated condition (many of which are presumably associated through a shared HLA restricted predisposition). It manifests as an intensely pruritic, blistering rash and diagnosis is dependent on the demonstration of IgA deposits in uninvolved skin. Although only 10% of patients have gastrointestinal symptoms, all will have some degree of enteropathy, be it villous atrophy or more subtle changes such as an increase in intraepithelial lymphocyte count or epithelial γ/δ T-cells.42 Treatment with a GFD takes about 6 months to show a response and 2 years for complete healing of the rash, which can be controlled in the meantime with dapsone.

DIAGNOSTIC CRITERIA

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. EPIDEMIOLOGY
  5. GLUTEN
  6. GENETIC PREDISPOSITION
  7. CLINICAL FEATURES AND DIAGNOSIS
  8. DIAGNOSTIC CRITERIA
  9. SEROLOGICAL MARKERS
  10. TREATMENT
  11. NONRESPONSIVE COELIAC DISEASE
  12. COMPLICATIONS OF CD
  13. FOLLOW-UP
  14. References

The criteria required for a diagnosis of CD have changed at various times over the last 30 years as our knowledge of the disease has improved. Although for purposes of research most will use the so-called ‘revised ESPGAN (European Society of Paediatric Gastroenterology and Nutrition) criteria’ (see below), as recently as 1996, at the Seventh International Symposium on CD, there was no general consensus on diagnostic criteria from among a group of leaders in the field from around the world.43

The original ESPGAN criteria of 197044 were essentially formulated for diagnosis in children and rely on a total of three small intestinal biopsies. In summary the requirements are:

• abnormal small intestinal mucosa while taking a gluten-containing diet;

• an unequivocal improvement in villous architecture on a GFD;

• deterioration of biopsy appearances with gluten challenge.

These criteria had a radical update in 1990,45 the revised ESPGAN criteria, and in essence require a single biopsy showing the characteristic CD lesion with an unequivocal clinical response on a GFD, although certain circumstances (e.g. on GFD at time of biopsy, adequacy of response to GFD, etc.) would mandate further biopsies. In addition, positive serology (anti-gliadin, anti-reticulin, anti-endomysium) at time of biopsy, with reversal on a GFD, can add weight to the diagnosis. Some would contend that in children the original ESPGAN criteria should still be fullfilled, especially if the child was young at the time of the initial biopsy. One of the main areas of diagnostic difficulty occurs in children less than 2 years of age, where other causes of enteropathy (e.g. cow’s milk enteropathy, transient gluten intolerance, post-enteritis syndrome, giardiasis) are often seen. These conditions, with the exception of giardiasis, are hardly ever seen in children older than 2 years or in adults. However, this review is concerned mainly with adult disease, as the authors are not paediatricians (although adult gastroenterologists will obviously see a number of patients in whom a diagnosis of CD was made in infancy and thus need to be aware of the pitfalls), and we would not wish to comment further except to say that the route to diagnosis in an individual child will depend on a variety of circumstances including age.

As for adult patients, we believe, as do others,46 that two biopsies are required to be certain about the diagnosis. Histological evidence of gluten sensitivity forms the core of most, if not all, definitions of CD and thus a demonstration of a change in the small intestinal mucosa on gluten-containing and gluten-free diets is the ideal for unequivocal diagnosis. It is recognized, however, that some physicians will be content with an initial characteristic biopsy on a gluten-containing diet, followed by clinical response to a GFD. CD still remains, however, a histological diagnosis and it is increasingly recognized that the histological definition is widening and can include patients with minimal changes on biopsy, such as an increase in intraepithelial lymphocytes (IEL).11 Likewise the clinical manifestations vary widely, including asymptomatic individuals, and before a lifelong, socially disabling diet is recommended, a firm diagnosis should be secured.

Problems can arise with this approach as with any other. There is currently no consensus on the best interval between biopsies, or the degree of histological improvement that is required. Three to 6 months is the usual interval in clinical practice, but it should be pointed out that the histological response may lag well behind the clinical response which is usually rapid, often occurring within days or weeks of gluten withdrawal.47, 48 Particularly in adults, complete mucosal healing may take up to 2 years or may even remain incomplete.49

In pragmatic everyday clinical practice these potential problems are not often important. It is only the occasional patient, e.g. where the initial biopsy was done on a GFD or is found in retrospect to be inadequate, who presents difficulty and in whom gluten challenge may be required. There is currently no standard for this but it is our practice to ask the patient to eat four slices of normal bread per day for 2 weeks, prior to a repeat biopsy, which should be brought forward if the patient becomes markedly symptomatic. Some patients however, may take much longer than this to show a relapse, particularly if they have previously been on a GFD for a long time. If there is still doubt it is reasonable to follow the patient over the course of a year or more, on a normal diet, and check for the appearance of anti-endomysial antibodies (AEA) as a very specific marker of mucosal relapse (see below), at which time a further biopsy can be obtained.

SEROLOGICAL MARKERS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. EPIDEMIOLOGY
  5. GLUTEN
  6. GENETIC PREDISPOSITION
  7. CLINICAL FEATURES AND DIAGNOSIS
  8. DIAGNOSTIC CRITERIA
  9. SEROLOGICAL MARKERS
  10. TREATMENT
  11. NONRESPONSIVE COELIAC DISEASE
  12. COMPLICATIONS OF CD
  13. FOLLOW-UP
  14. References

The advent of very specific serological markers for CD has already altered the epidemiology of the disease, as more and more people who previously would not have been considered for biopsy, are now identified as being at risk, and may yet make a significant impact on actual diagnosis, which currently still depends on small intestinal biopsy as the gold standard. The main role of these tests in clinical practice is in screening patients who may have non-specific symptoms or a condition associated with CD (e.g. IDDM) in order to avoid an unnecessary biopsy, which should probably still be the first investigation in a patient strongly suspected of having the disease. They can also be used to assess dietary compliance.

Table 3 shows the four currently available tests, and compares two main studies, in adults50 and children,51 which directly compared them all. As can be seen, with the very high levels of sensitivity and specificity, IgA AEA is the best serological test for use in CD diagnosis, but does have three disadvantages. Firstly the assay technique, indirect immunofluorescence, is subject to inter-operator variation and so reliability of any particular laboratory must be established. Secondly, it is of no use in the 3–10% of CD patients who have associated IgA deficiency and for this reason combining it with an estimation of immunoglobulins or IgG AGA may be the best approach. Lastly the assay substrate is usually monkey oesophagus, although recently human umbilical cord has been shown to give improved results,52 and has the added benefits of being more ethically acceptable and much cheaper.

Table 3.  . Comparison of two studies using the four currently available tests Thumbnail image of

Although AGAs (anti-gliadin antibodies) are less sensitive and specific than AEA they have the advantage of an ELISA assay, being therefore easier and quicker to undertake and with much less inter-laboratory variation. Unlike AEA they are not disease specific, being found in a number of other conditions as well as ‘normal’ controls.

ARA is very similar in most respects to AEA but because of the reduced sensitivity, AEA is to be preferred.

These antibodies can be used to assess dietary compliance, but it is unclear which is best for this purpose.

Recently the enzyme tissue transglutaminase (tTG) has been found to be the antigen for AEA,53 and there is speculation that this enzyme may have a direct role in disease pathogenesis. Molberg et al.54 have shown that it may be involved in the generation of gliadin epitopes, with marked increased binding affinity for DQ2 and enhanced capacity for T-cell stimulation, by selective deamidation of specific glutamine residues (found in very high proportion in gluten) to glutamic acid.

Our group, amongst others, is working on an ELISA assay for anti-tTG antibodies.

TREATMENT

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. EPIDEMIOLOGY
  5. GLUTEN
  6. GENETIC PREDISPOSITION
  7. CLINICAL FEATURES AND DIAGNOSIS
  8. DIAGNOSTIC CRITERIA
  9. SEROLOGICAL MARKERS
  10. TREATMENT
  11. NONRESPONSIVE COELIAC DISEASE
  12. COMPLICATIONS OF CD
  13. FOLLOW-UP
  14. References

GFD

Once a firm diagnosis of CD has been established the cornerstone of treatment is the gluten-free diet (GFD). In theory this is straightforward and some would dismiss, erroneously, the importance of continued contact with patients in the clinic once the diagnosis has been made and the advice to ‘avoid gluten’ has been dispensed. In practice, of course, things are not so easy. As with any disease the newly diagnosed patient needs information. A careful explanation of the disease and the toxicity of a ubiquitous foodstuff, of which patients are invariably very fond, is paramount. This is particularly important in patients with very mild or absent symptoms. The avoidance of future ill health, or reversal of current problems, including anaemia, osteoporosis, depression and infertility needs to be explained. Whether or not the increased frequency of malignancy, and in particular small intestinal lymphoma, in the untreated patient is discussed is debatable and should be decided with discretion on a patient by patient basis. The aim should not be to frighten the patient into dietary compliance, but to equip him or her with the necessary facts to be able to decide for themselves. However, in a time of increasing litigation against doctors, failure to disclose information about a patient’s disorder (no matter how noble the intention) which may influence their perceived necessity for treatment compliance, may be hard to defend.

Patients (and doctors) are often confused about which foods are, or are not, gluten-free. It is essential therefore that an experienced dietitian spends time going over the dietary modifications that are required. Joining the Coeliac Society (PO Box, 220, High Wycombe, Bucks, UK), who publish lists of gluten-free products, is also to be strongly advised. There is now a wide range of gluten-free breads, biscuits, ‘pasta’, etc. made by a variety of companies and most of these can be obtained on FP 10 prescription if it is marked ‘ACBS’.

Some areas, such as beer, consistently cause confusion. Beer, of all types, was removed from the Coeliac Society’s list of safe foodstuffs after it became clear that it contains appreciable quantities of barley gluten prolamin (hordein).55, 56 Enzymatic destruction of gluten/hordein during brewing was previously assumed to occur. Other potential problem areas are still coming to light. For example, significant quantities of hordein have been found in cornflakes,57 widely thought to be gluten-free, owing to the malt flavouring used. We have four patients in whom consumption of cornflakes is likely to have been the cause of relapse or initial failure to respond on a GFD. Such problems may well be negligible for the majority of patients (who vary in their sensitivity to gluten concentration) but should be borne in mind in those who do develop difficulties as the commonest cause of relapse is gluten consumption, inadvertent or otherwise. The potential problem of oats has been discussed above.

Nutrient deficiencies

At the time of initial investigation and diagnosis patients will have had a variety of routine blood tests including FBC, biochemical profile and, hopefully, iron and folate studies. Supplements to replace the latter two, if deficient, may not be required routinely, correcting relatively quickly on a GFD. On occasion, however, a deficiency of calcium will require calcium and vitamin D supplementation, and rarely patients are hospitalized with tetany requiring urgent i.v. replacement. Similarly life-threatening hypokalaemia or hypomagnesaemia can rarely occur.

The treatment of other aspects of CD is discussed below.

NONRESPONSIVE COELIAC DISEASE

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. EPIDEMIOLOGY
  5. GLUTEN
  6. GENETIC PREDISPOSITION
  7. CLINICAL FEATURES AND DIAGNOSIS
  8. DIAGNOSTIC CRITERIA
  9. SEROLOGICAL MARKERS
  10. TREATMENT
  11. NONRESPONSIVE COELIAC DISEASE
  12. COMPLICATIONS OF CD
  13. FOLLOW-UP
  14. References

This occurs when a patient who has previously been well on a GFD relapses both clinically and histologically, or, following diagnosis, fails to respond to a gluten-free diet (which assumes that the diagnosis is correct). The management of such patients has been dealt with in an excellent review by O’Mahony et al. 58

The commonest cause of non-responsiveness is continued gluten intake but the most important things to exclude are small intestinal lymphoma and ulcerative jejunitis, both of which are discussed below. If such diagnoses can be excluded, and the patient is genuinely non-responsive to gluten withdrawal, bearing in mind that histological response often trails well behind clinical response, then a number of therapeutic options can be tried,58 although little hard evidence exists of clear-cut benefit for any of these. Steroids have been found in several studies to be of benefit in patients on a gluten-containing diet,59, 60 but should only be considered in patients non-resposive to GFD, in whom other causes have been excluded, or in those with complicating ulcerative jejunitis,58, 61 (but see also below as there can be problems with this). Azathioprine has also been used.62 There is no evidence for the use of cyclosporine and in our experience it does not work. Total parenteral nutrition may occasionally be required in non-responsive patients with continued weight loss.

COMPLICATIONS OF CD

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. EPIDEMIOLOGY
  5. GLUTEN
  6. GENETIC PREDISPOSITION
  7. CLINICAL FEATURES AND DIAGNOSIS
  8. DIAGNOSTIC CRITERIA
  9. SEROLOGICAL MARKERS
  10. TREATMENT
  11. NONRESPONSIVE COELIAC DISEASE
  12. COMPLICATIONS OF CD
  13. FOLLOW-UP
  14. References

Malignancy

The overall mortality in CD is approximately twice that of the general population, with the increase occurring predominantly within the first year after diagnosis.63 The excess deaths are due mainly to malignancies (Table 4), with intestinal lymphoma accounting for the majority of these.

The association of steatorrhoea with lymphoma has been noted for many years64 with the early reports concluding that the tumour caused the malabsorption. Gough et al.65 in 1962, suggested that lymphoma was a complication of CD and this has since been well established in a number of series and reports. However, the exact prevalence of lymphoma in CD is difficult to determine accurately, partly because it can arise in patients with silent or previously undiagnosed CD—the occurrence of which makes it difficult to know the exact prevalence of CD itself. Thus if lymphoma is the presenting clinical episode the presence of concurrent CD may not be sought.

Several studies have gone a long way in answering this question though. In a mortality study of 653 patients with CD, 17 deaths were attributed to lymphoproliferative disease where 0.55 were expected, giving a relative risk of 31.63 This study also showed a three-fold increase in all malignant disease. Holmes et al.41 looking specifically at malignancy in a population of 210 CD patients, found a two-fold risk of developing cancer, with a relative risk of developing non-Hodgkin’s lymphoma of about 40. The importance of this latter study is that malignant complications of CD were correlated with GFD. Thus for patients who had taken a GFD for 5 years or more the risk of cancer was not increased above the general population. For those taking a reduced gluten or normal diet the relative risk of lymphoma was 77.8 and for cancers of the mouth, pharynx and oesophagus was 22.7. The protective effect of the GFD can also be inferred from the finding by Logan et al.63 that the mortality rate in patients diagnosed with CD in childhood was similar to the general population.

Histologically the lymphoma complicating CD was originally thought to be of monocyte/macrophage origin, sharing various markers with these cell types, and labelled as ‘malignant histiocytosis of the intestine’.66, 67 It has since been demonstrated that it is of T-cell origin,68 which fits well with the abundant data suggesting that CD is a T-cell driven disorder, and is referred to as ‘enteropathy associated T-cell lymphoma’ (EATCL).

From a clinical perspective it is well recognized that it can be very difficult to diagnose EATCL, and as the disease is often widespread at presentation the prognosis is usually very poor. It will commonly present as an acute abdominal emergency with obstruction or perforation, perhaps preceeded by abdominal pain and weight loss,69 and will often arise in a patient not previously known to have CD. Alternatively, a patient with known CD may become unresponsive to their GFD with a recurrence of symptoms such as diarrhoea, weight loss and lethargy. Profound muscle weakness can also be a prominent feature.70 Although inadvertent gluten consumption is commonly the cause of such secondary non-responsiveness, this cannot be assumed, and in addition to small intestinal biopsies (for which push enteroscopy may prove useful) small bowel contrast radiology and CT scanning should be undertaken if lymphoma is suspected. Some authorities recommend laparotomy, if lymphoma is strongly suspected and hard evidence is lacking,70 or laparoscopy.71

Treatment of EATCL is unsatisfactory with a low survival rate. Recent data from Ireland, by Egan et al. has addressed this problem and found an overall 1 year survival rate of 31% and at 5 years of 11%.71 In those patients with a previous diagnosis of CD and diagnosed in life with lymphoma the mean survival was only 2.25 months. The majority of patients in this study had CD and lymphoma diagnosed during the same clinical episode. Surgery, chemotherapy and radiotherapy may all be used depending on the circumstances. If confined to the intestine resection may result in cure. However, in the study of Egan et al. long-term survival was seen almost exclusively in those treated with chemotherapy.

After lymphoma the commonest invasive malignancy is adenocarcinoma of the intestine. Swinson et al.72 found a relative risk of 82.6 amongst CD patients compared to the general population, but as they point out this is anyway a very rare tumour so the risk to an individual CD patient is small. Presentation can take a variety of forms but anaemia and gastrointestinal blood loss are prominent features along with weight loss, obstruction, vomiting and abdominal pain.70 Radiology is more likely to be helpful in diagnosis than it is with lymphoma and long-term survival is possible if diagnosed early and the tumour is amenable to resection. Whereas a GFD is protective against lymphoma development, a recent case report suggests that it is not protective against development of small intestinal adenocarcinoma.73 The authors of this report suggest that the predisposition to this tumour in CD is probably genetic. However, it is difficult to be clear about the situation from one case.

Ulcerative jejunitis

A certain amount of confusion has surrounded this rare complication of CD which is reflected by the variety of names by which it has been labelled, including: idiopathic chronic ulcerative enteritis,74 chronic ulcerative non-granulomatous jejunitis,75 and chronic non-specific ulcerative duodenojejunoileitis.76 One of the main difficulties has been deciding whether all cases are associated with CD or if some represent a separate disorder.

The ulcers tend to be transverse and located mainly in the jejunum or ileum. Scarring can lead to strictures with intervening areas of dilated bowel. Histologically there is a variable thickness of ulceration, but often extending to the muscularis, with oedema, muscular hypertrophy and fibrosis and a mixed inflammatory infiltrate in the ulcer bases.74, 77 There is a variable degree of villous atrophy in adjacent and distant mucosa, but normal mucosa can be present.74

Mills et al. reported five cases and reviewed a further 27 from the English literature.74 Severe malabsorption with multiple nutrient deficiencies were characteristic with diarrhoea and weight loss being almost invariable. Finger clubbing was prominent in the five cases reported, perhaps revealing the length of illness in these patients. Eleven cases were said to show a response to a GFD but CD was confirmed in only three, based on the criteria for diagnosis at that time, requiring a recovery of villous height. By the current ESPGAN criteria all these patients could possibly be considered to have CD, having apparently shown clinical response to a GFD, although demonstration of mucosal healing is ideal. It is not clear, however, for how long the GFDs were adhered to, or what the intervals between biopsies were.

Baer et al.77 reviewed 47 cases and considered 22 of them to have underlying CD, although their criteria for inclusion in this group were considerably more lax than those of Mills et al.74 A further 11 patients were categorized as ‘unclassified sprue’. The authors felt that some if not most of this latter group also had CD but with no response to GFD either because of the ulceration or their advanced age. This is perhaps not unreasonable as ulcerative jejunitis (UJ) is one of the differential diagnoses for non-responsive CD.58 This situation is highlighted in a very recent case report from Italy,78 in which a middle-aged woman with ulcerative jejunitis and subtotal villous atrophy on duodenal biopsy failed initially to respond to a GFD. She was however HLA-DQ2 positive and a son was diagnosed with silent CD on the basis of positive serology and subsequent flat biopsy. The patient was treated surgically and with a strict GFD and steroids. A repeat small intestinal biopsy showed regrowth of villi. This report also demonstrates that HLA typing, use of AEA and screening of family members may be helpful in diagnosis. Interestingly AEA had been negative in the patient concerned, although was performed after 7 months of GFD, despite lack of clinical response.

The series of Mills et al. and Baer et al. both reveal a high mortality rate with death often following haemorrhage, perforation or obstruction and often on a background of malnutrition. This emphasizes the need for an aggressive approach to diagnosis and treatment. Diagnosis can be very difficult with small bowel radiology often being unhelpful unless strictures are present. In many cases to date diagnosis was only made at laparotomy with full thickness biopsy/resection. Push enteroscopy may prove useful in this regard.79

Surgical resection of the ulcers, especially if localized to one part of the intestine, can be curative. A novel intervention for this condition is strictureoplasty and was used, probably for the first time, in the case referred to above.79 A strict GFD should be instituted, even if it is unclear whether or not there is underlying CD, as discussed. Steroids have often been used, sometimes with benefit, but care should be taken because in the series of Baer et al.77 intestinal perforation occurred in 10 patients while on steroids. Azathioprine has also been used.80

Whether ulcerative jejunitis represents a manifestation of EATCL, as has been suggested,81 is not resolved. It has recently been shown, though, that T-cell monoclonality is a feature of the ulcers both in UJ and EATCL.82

Disorders of bone metabolism

Abnormalities of bone metabolism have long been noted in CD.83 Osteomalacia is well recognized,47 even in the absence of abdominal symptoms, and responds to calcium and vitamin D supplementation. Bone pain, pseudofractures or deformity may be features but it is often asymptomatic and found following a raised serum alkaline phosphatase level with calcium and phosphate levels often being normal.84 Vitamin D (25(OH) D3) and parathyroid hormone levels are useful, with bone biopsy giving absolute confirmation, although this is not now commonly performed.

In recent years it has become increasingly recognized that osteopenia and osteoporosis are very common features of adult CD,85[86]–87 although the exact mechanisms are unclear.88 Bone mineral density (BMD) is almost always reduced in untreated CD but has also been found with very high prevalence in patients on a GFD. A study of 65 patients already on a GFD found ‘osteoporosis’ in 50% of the men and 47% of the women.89 The definition of osteoporosis was a T score of less than –2 on DEXA (dual energy X-ray absorptiometry) scanning whereas less than – 2.5 (i.e. 2.5 s.d. below mean peak value in young adults) is a more widely recognized definition.90 Nevertheless this is a very high frequency of osteopenia.

From a different perspective, the rate of undetected CD has been found to be 10-fold higher than in the normal population, when screening patients wtih clinical osteoporosis with IgA anti-gliadin antibodies and small intestinal biopsies in positives.91

An important point to note is that symptoms of malabsorption do not correlate with bone disease. Amongst a group of asymptomatic, treated CD patients, it was found that those with reduced BMD could not be predicted clinically but were more likely to have subtotal or partial villous atrophy on biopsy.86 In another study no correlation between BMD and severity of clinical or biochemical abnormalities was found.92

Although one early study did not show any improvement in BMD after GFD93 many subsequent studies have.94[95]–96 Valdimarson et al. found a significant increase in BMD at all sites measured in a group of 63 newly diagnosed CD patients after 1 year of GFD.94 Significant increase in BMD with GFD was confirmed in another study, although after 1 year BMD was still lower in the patients compared to controls.95 It is doubtful whether bone mineralization continues to increase after the initial improvement found after the first year of GFD. In a recent study of 25 newly treated CD patients BMD was measured at diagnosis and after a mean of 37 months of treatment.97 There was a significant increase in BMD in the majority of patients but only three achieved normal bone mass. The authors also concluded from analysis of a small group of patients who were followed at different intervals of time that the increase in bone mineralization was greatest during the first year of treatment. In contrast, in an earlier study no difference in BMD was found between CD patients responsive to diet, and whose intestinal mucosa had normalized at least 4 years previously, and healthy controls.98

It seems that in children and adolescents that although BMD can be significantly reduced at diagnosis, GFD leads to complete recovery of bone mass emphasizing the importance of early diagnosis and strict adherence to a GFD.92, 99

Osteoporosis carries significant fracture risk with increased morbidity, mortality and attendant socioeconomic implications. Thus screening of CD patients is important and now recommended.87, 88, 100 This can be easily and relatively cheaply undertaken by DEXA scanning. If osteoporosis is found, besides strict adherence to a GFD and consideration of repeat biopsy in those already treated, treatment may comprise HRT in post-menopausal women, biphosphanates or calcitonin and is reviewed elsewhere, both in general,101 and specifically with reference to post-menopausal women.102

For CD patients with osteopenia but who are not actually osteoporotic it is less clear what action should be taken. Repeat biopsy should be considered and strict adherence to GFD emphasized. Dietary calcium supplementation up to 1500 mg/day has been recommended,88, 100 particularly for women who at menopause will suffer large bone losses. Smoking should be discouraged and exercise advised. Monitoring by repeat DEXA scan, e.g. at 1 year, is also important and allows an estimate of rate of change of BMD.

Splenic atrophy

This is considered to be a relatively common complication of adult CD,103, 104 although the mechanism of its occurrence is unknown. The finding of blood film features of hyposplenism (e.g. Howell–Jolly bodies), with no obvious cause, should perhaps prompt a search for CD.105

Although partially reversible by GFD106 it may still predispose to fatal bacterial infections. It has been suggested that in some situations, e.g. malnutrition, surgery, that pneumoccocal vaccination be administered.107 Whether this should be advocated for all CD patients with evidence of hyposplenism is not known, but infections should obviously be treated without delay.

FOLLOW-UP

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. EPIDEMIOLOGY
  5. GLUTEN
  6. GENETIC PREDISPOSITION
  7. CLINICAL FEATURES AND DIAGNOSIS
  8. DIAGNOSTIC CRITERIA
  9. SEROLOGICAL MARKERS
  10. TREATMENT
  11. NONRESPONSIVE COELIAC DISEASE
  12. COMPLICATIONS OF CD
  13. FOLLOW-UP
  14. References

Once the initial phase of biopsies, diagnosis and initiation of the GFD has been dealt with the question of who should review CD patients—GP or gastroenterologist—and how frequently, often arises. Some contend that since the vast majority of patients remain well with strict adherence to a GFD, follow-up should be undertaken by the patient’s GP, with, for example, yearly weight, FBC, folate, calcium and alkaline phosphatase.108 Certainly this is one way of doing things, especially if there is good liaison between GP and local gastroenterologist and referral is made early for any problems arising such as weight loss, new abdominal pain, anaemia, etc. What is clear is that follow-up should be lifelong.32, 48 This allows reinforcement of the continuing need for strict adherence to the GFD, and the early detection of any problems.

It is our experience, however, that a significant proportion of patients may run into problems. There are an increasing number of conditions recognized to be associated with CD and in addition new tests (e.g. AEA) at our disposal require some experience to utilize appropriately. Screening of first degree relatives is also an issue. On the one hand CD may appear to be a deceptively straightforward condition but on the other hand can prove to harbour some complex problems, both in diagnosis and treatment. It is our belief therefore that follow-up should be at a gastroenterology clinic.

References

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. EPIDEMIOLOGY
  5. GLUTEN
  6. GENETIC PREDISPOSITION
  7. CLINICAL FEATURES AND DIAGNOSIS
  8. DIAGNOSTIC CRITERIA
  9. SEROLOGICAL MARKERS
  10. TREATMENT
  11. NONRESPONSIVE COELIAC DISEASE
  12. COMPLICATIONS OF CD
  13. FOLLOW-UP
  14. References
  • 1
    Visakorpi JK. Changing features of coeliac disease. In: Mäki M, Collin P, Visakorpi JK, eds. Coeliac Disease. Tampere: Coeliac disease study group, 1997: 1–7.
  • 2
    Ascher H, Krantz I, Kristiansson B. Increasing incidence of coeliac disease in Sweden. Arch Dis Child 1991; 66: 608 11.
  • 3
    Mäki M, Holm K, Ascher H, Greco L. Factors affecting clinical presentation of coeliac disease: Role of type and amount of gluten-containing cereals in the diet. In: Auricchio S, Visakorpi JK, eds. Common Food Intolerances 1: Epidemiology of Coeliac Disease. Basel: Karger, 1992: 76–82.
  • 4
    Ascher H, Krantz I, Rydberg L, Nordin P, Kristiansson B. Influence of infant feeding and gluten intake on coeliac disease. Arch Dis Child 1997; 76: 113 17.
  • 5
    Catassi C, Fabiani E, Rätsch IM, et al. The coeliac iceberg in Italy. A multicentre antigliadin antibodies screening for coeliac disease in school-age subjects. Acta Paediatr 1996; 85(Suppl. 412): 29 35.
  • 6
    Grodzinsky E, Franzen L, Hed J, Ström M. High prevalence of celiac disease in healthy adults revealed by antigliadin antibodies. Ann Allergy 1992; 69: 66 70.
  • 7
    Fasano A. Where have all the American celiacs gone? Acta Paediatr 1996; 85(Suppl. 412): 20 4.
  • 8
    Johnston SD, Watson RGP, McMillan SA, et al. Preliminary results from follow-up of a large scale population survey of antibodies to gliadin, reticulin and endomysium. Acta Paediatr 1996; 85(Suppl. 412): 61 4.
  • 9
    Catassi C, Rätsch I-M, Fabiani E, et al. Coeliac disease in the year 2000: exploring the iceberg. Lancet 1994; 343: 200 3.
  • 10
    Mäki M & Collin P. Coeliac disease. Lancet 1997; 349: 1755 59.
  • 11
    Ferguson A, Arranz E, O’Mahony S. Clinical and pathological spectrum of coeliac disease—active, silent, latent, potential. Gut 1993; 34: 150 1.
  • 12
    Gee S. On the coeliac disease. St Bart Hosp Rep 1888; 24: 17 20.
  • 13
    Dicke WK. Coeliakie. MD Thesis, Utrecht, 1950.
  • 14
    Dicke WK, Weijers HA, Van De Kamer JH. Coeliac disease. II. The presence in wheat of a factor having a deleterious effect in cases of coeliac disease. Acta Paediatr Scand 1953; 42: 34 42.
  • 15
    Van De Kamer JH, Weijers HA, Dicke WK. Coeliac disease. IV. An investigation into the injurious constituents of wheat in connection with their action on patients with coeliac disease. Acta Paediatr Scand 1953; 42: 223 31.
  • 16
    Janatuinen EK, Pikkarainen PH, Kemppainen TA, et al. A comparison of diets with and without oats in adults with celiac disease. N Engl J Med 1995; 333: 1033 37.
  • 17
    Srinivasan U, Leonard N, Jones E, et al. Absence of oats toxicity in adult coeliac disease. Br Med J 1996; 313: 1300 1.
  • 18
    Hardman CM, Garioch JJ, Leonard JN, et al. Absence of toxicity of oats in patients with dermatitis herpetiformis. N Engl J Med 1997; 337: 1884 7.
  • 19
    Sturgess R, Day P, Ellis HJ, et al. Wheat peptide challenge in coeliac disease. Lancet 1994; 343: 758 61.
  • 20
    Shidrawi RG, Day P, Przemioslo R, Ellis HJ, Nelufer JM, Ciclitira PJ. In vitro toxicity of gluten peptides in coeliac disease assessed by organ culture. Scand J Gastroenterol 1995; 30: 758–62.
  • 21
    Parnell NDJ, Ellis HJ, Ciclitira P. Absence of toxicity of oats in patients with dermatitis herpetiformis. N Engl J Med 1998; 338: 1470.
  • 22
    Sollid L, Markussen G, Ek J, Gjerde H, Vartdal F, Thorsby E. Evidence for a primary association of celiac disease to a particular HLA-DQ α/β heterodimer. J Exp Med 1989; 169: 345 50.
  • 23
    Tighe MR & Ciclitira PJ. The gluten–host interaction. In: Howdle PD, ed. Coeliac Disease. Bailliere’s Clinical Gastroenterology 1995; 9: 2: 211–30.
  • 24
    Marsh M, Bjarnason I, Shaw J, et al. Studies of intestinal lymphoid tissue. XIV-HLA status, mucosal morphology, permeability and epithelial lymphocyte populations in first degree relatives of patients with coeliac disease. Gut 1990; 31: 32 6.
  • 25
    Walker-Smith JA. Celiac disease. In: Walker WA, Durie PR, Hamilton JR, Walker-Smith JA, eds. Pediatric Intestinal Disease, Vol 2. Philadelphia: BC Decker, 1991: 700–18.
  • 26
    Greco L, Corazza G, Babron MC, et al. Genome search in celiac disease. Am J Human Genet 1998; 62: 669 75.
  • 27
    Lundin K, Scott H, Hansen T. Gliadin specific HLA-DQ (A1*0501, B1*0201) restricted T cells isolated from the small intestinal mucosa of coeliac patients. J Exp Med 1993; 178: 187 96.
  • 28
    Nilsen EM, Lundin KEA, Krajci P, et al. Gluten specific, HLA-DQ restricted T cells from coeliac mucosa produce cytokines with Th1 or Th0 profile dominated by interferonγ. Gut 1995; 37: 766 76.
  • 29
    Johansen BH, Gjertsen HA, Vartdal H, et al. Binding of peptides from the N-terminal region of α-gliadin to the celiac disease-associated HLA-DQ2 molecule assessed in biochemical and T-cell assays. Clin Immunol Immunopathol 1996; 79: 288 93.
  • 30
    Shidrawi RG, Parnell NDJ, Ciclitira PJ, Travers P, Evan G, Rosen-Bronson S. Binding of gluten-derived peptides to the HLA-DQ2 (α1*0501, β1*0201) molecule, assessed in a cellular assay. Clin Exp Immunol 1998; 111: 158 65.
  • 31
    Gjertsen HA, Lundin KEA, Sollid LM, Eriksen JA, Thorsby E. T cells recognise a peptide derived from α-gliadin presented by the celiac disease-associated HLA-DQ (α1*0501, β1*0201) heterodimer. Human Immunol 1994; 39: 243 52.
  • 32
    Corazza GR & Gasbarrini G. Coeliac disease in adults. In: Howdle PD, ed. Coeliac Disease. Bailliere’s Clinical Gastroenterology 1995; 9: 2: 329–50.
  • 33
    Corazza GR, Frisoni M, Treggiari EA, et al. Subclinical celiac sprue: increasing occurrence and clues to its diagnosis. J Clin Gastroenterol 1993; 16: 16 21.
  • 34
    Thain ME, Hamilton JR, Erlich RM. Coexistence of diabetes mellitus and celiac disease. J Pediatr 1974; 85: 527 9.
  • 35
    Savilahti E, Simell O, Koskimies S, et al. Celiac disease in insulin-dependent diabetes mellitus. J Pediatr 1986; 108: 690 3.
  • 36
    Collin P, Reunala T, Pukkala E, et al. Coeliac disease—associated disorders and survival. Gut 1994; 35: 1215 18.
  • 37
    Chapman RWG, Laidlow JM, Colin-Jones D, Eade OE, Smith CL. Increased prevalence of epilepsy in coeliac disease. Br Med J 1978; 22: 250 1.
  • 38
    Holmes GKT. Non-malignant complications of coeliac disease. Acta Paediatr 1996; 412(Suppl.): 68 75.
  • 39
    Gobbi G, Bouquet F, Greco L, et al. Coeliac disease, epilepsy, and cerebral calcifications. Lancet 1992; 340: 439 43.
  • 40
    Cronin CC, Jackson LM, Feighery C, et al. Coeliac disease and epilepsy. Q J Med 1998; 91: 303 8.
  • 41
    Holmes GKT, Prior P, Lane MR, Pope D, Allan RN. Malignancy in coeliac disease—effect of a gluten free diet. Gut 1989; 30: 333 8.
  • 42
    Fry L. Dermatitis herpetiformis: from gut to skin, contributions to the nature of coeliac disease. In: Mäki M, Collin P, Visakorpi JK, eds. Coeliac Disease. Tampere: Coeliac disease study group, 1997: 67–74.
  • 43
    Walker-Smith JA. Discussion of diagnostic criteria for coeliac disease at Tampere meeting. In: Mäki M, Collin P, Visakorpi JK, eds. Coeliac Disease. Tampere: Coeliac disease study group, 1997: 191–3.
  • 44
    Meeuwisse GW. Diagnostic criteria in coeliac disease. Acta Paediatr Scand 1970; 59: 461 3.
  • 45
    Walker-Smith JA, Guandalini S, Schmitz J, Shmerling DH, Visakorpi JK. Revised criteria for the diagnosis of coeliac disease. Report from a working group. Arch Dis Child 1990; 65: 909 11.
  • 46
    Howdle PD. Two biopsies for adults. In: Mäki M, Collin P, Visakorpi JK, eds. Coeliac Disease. Tampere: Coeliac disease study group, 1997: 185–8.
  • 47
    Trier JS. Celiac sprue. N Engl J Med 1991; 325: 1373 4.
  • 48
    Howdle PD & Losowsky MS. Coeliac disease in adults. In: Marsh MN, ed. Coeliac Disease. Oxford: Blackwell Scientific Publications, 1992: 49–80.
  • 49
    Grefte JMM, Bouman JG, Grond J, Jansen W, Kleibeuker JH. Slow and incomplete histological and functional recovery in adult gluten sensitive enteropathy. J Clin Pathol 1988; 41: 886 91.
  • 50
    Ferreira M, Lloyd Davies S, Butler M, Scott D, Clark M, Kumar P. Endomysial antibody: is it the best screening test for coeliac disease? Gut 1992; 33: 1633 7.
  • 51
    Lerner A, Kumar V, Iancu TC. Immunological diagnosis of childhood coeliac disease: comparison between antigliadin, antireticulin and antiendomysial antibodies. Clin Exp Immunol 1994; 95: 78 82.
  • 52
    Ladinser B, Rossipal E, Pittschieler K. Endomysium antibodies in coeliac disease: an improved method. Gut 1994; 35: 776 8.
  • 53
    Dieterich W, Ehnis T, Bauer M, et al. Identification of tissue transglutaminase as the autoantigen of celiac disease. Nat Med 1997; 3: 797 801.
  • 54
    Molberg Ø, McAdam SN, Körner R, et al. Tissue transglutaminase selectively modifies gliadin peptides that are recognized by gut-derived T cells in celiac disease. Nat Med 1998; 4: 713 17.
  • 55
    Ellis HJ, Freedman AR, Ciclitira PJ. Detection and estimation of the barley prolamin content of beer and malt to assess their suitability for patients with coeliac disease. Clin Chim Acta 1990; 189: 123 30.
  • 56
    Ellis HJ, Doyle AP, Day P, Wieser H, Ciclitira PJ. Demonstration of the presence of coeliac-activating gliadin-like epitopes in malted barley. Int Arch Allergy Immunol 1994; 104: 308 10.
  • 57
    Ellis HJ, Parnell NDJ, Ciclitira PJ. Cornflakes and coeliac disease. Gut 1998; 42(Suppl. 1): A35(Abstract).
  • 58
    O’Mahony S, Howdle PD, Losowsky MS. Review article: management of patients with non-responsive coeliac disease. Aliment Pharmacol Ther 1996; 10: 671 80.
  • 59
    Adlersberg D, Colcher H, Drachman SR. Studies on the effects of cortisone pituitary adrenocorticotrophic hormone (ACTH) in the sprue syndrome. Gastroenterology 1951; 19: 674 97.
  • 60
    Wall AJ, Douglas AP, Booth CC, Pearse AGE. Response of the jejunal mucosa in adult coeliac disease to oral prednisolone. Gut 1970; 11: 7 14.
  • 61
    Jones PE & Gleeson MH. Mucosal ulceration and mesenteric lymphadenopathy in coeliac disease. Br Med J 1973; 3: 212 13.
  • 62
    Hamilton JD, Chambers RA, Wynn-Williams A. Role of gluten, prednisone, and azathioprine in non-responsive coeliac disease. Lancet 1976; 1: 1213 16.
  • 63
    Logan RFA, Rifkind EA, Turner ID, Ferguson A. Mortality in celiac disease. Gastroenterology 1989; 97: 265 71.
  • 64
    Fairley NH & Mackie FP. The clinical and biochemical syndrome in lymphadenoma and allied diseases involving the mesenteric glands. Br Med J 1937; 1: 375 80.
  • 65
    Gough KR, Read AE, Naish JM. Intestinal reticulosis as a complication of idiopathic steatorrhoea. Gut 1962; 3: 232 9.
  • 66
    Isaacson PG & Wright DH. Malignant histiocytosis of the intestine: its relationship to malabsorption and ulcerative jejunitis. Human Pathol 1978; 9: 661 77.
  • 67
    Isaacson PG, Jones DB, Sworn MJ, Wright DH. Malignant histiocytosis of the intestine: report of three cases with immunological and cytochemical analysis. J Clin Pathol 1982; 35: 510 16.
  • 68
    Isaacson PG, O’Connor NTJ, Spencer J, et al. Malignant histiocytosis of the intestine: a T cell lymphoma. Lancet 1985; ii: 688 91.
  • 69
    Wright DH. The major complications of coeliac disease. In: Howdle PD, ed. Coeliac Disease. Bailliere’s Clinical Gastroenterology 1995; 9: 2: 351–69.
  • 70
    Holmes GKT & Thompson H. Malignancy as a complication of coeliac disease. In: Marsh MN, ed. Coeliac Disease. Oxford: Blackwell Scientific Publications, 1992: 105–35.
  • 71
    Egan LJ, Walsh SV, Stevens FM, Connolly CE, Egan EL, McCarthy CF. Celiac-associated lymphoma. A single institution experience of 30 cases in the combination chemotherapy era. J Clin Gastroenterol 1995; 21: 123 9.
  • 72
    Swinson CM, Slavin G, Coles EC, Booth CC. Coeliac disease and malignancy. Lancet 1983 I. 111–15.
  • 73
    Kingham JG, Ramanaden D, Dawson A. Metachronous small-bowel adenocarcinoma in coeliac disease: Gluten-free diet is not protective. Scand J Gastroenterol 1998; 33: 218 22.
  • 74
    Mills PR, Brown IL, Watkinson G. Idiopoathic chronic ulcerative enteritis. Q J Med 1980; XLIX (194): 133–49.
  • 75
    Jeffries GH, Steinberg H, Sleisenger MH. Chronic ulcerative (nongranulomatous) jejunitis. Am J Med 1968; 44: 47 59.
  • 76
    Modigliani R, Poitras P, Galian A, et al. Chronic non-specific ulcerative duodenojejunoileitis: report of four cases. Gut 1979; 20: 318 28.
  • 77
    Baer AN, Bayless TM, Yardley JH. Intestinal ulceration and malabsorption syndromes. Gastroenterology 1980; 79: 754 65.
  • 78
    Biagi F, Poggioli G, Mazzoni G, Corazza GR. Case report: Intestinal strictures. Lancet, 1998; 352: 876.
  • 79
    Green JA, Barkin JS, Gregg PA, Kohen K. Ulcerative jejunitis in refractory celiac disease: enteroscopic visualization. Gastrointest Endosc 1993; 39: 584 5.
  • 80
    Enns R, Lay T, Bridges R. Use of azathioprine for nongranulomatous ulcerative jejunoileitis. Can J Gastroenterol 1997; 11: 503 6.
  • 81
    Isaacson P & Wright DH. Malignant histiocytosis of the intestine. Its relationship to malabsorption and ulcerative jejunitis. Human Pathol 1978; 9: 661 77.
  • 82
    Ashton-Key M, Diss TC, Pan L, Du MQ, Isaacson PG. Molecular analysis of T-cell clonality in ulcerative jejunitis and enteropathy-associated T-cell lymphoma. Am J Pathol 1997; 151: 493 8.
  • 83
    Bennett TI, Hunter D, Vaughan JM. Idiopathic steatorrhoea (Gee’s disease). A nutritional disturbance associated with tetany, osteomalacia and anaemia. Q J Med 1932; 1: 603 77.
  • 84
    Cooke WT & Holmes GKT. Clinical presentation. In: Coeliac Disease. Edinburgh: Churchill Livingstone, 1984: 90–5.
  • 85
    Corazza GR, Di Sario A, Ceccheti L, et al. Bone mass and metabolism in patients with celiac disease. Gastroenterology 1995; 109: 122 8.
  • 86
    Walters JR, Banks LM, Butcher GP, Fowler CR. Detection of low bone mineral density by dual energy X-ray absorptiometry in unsuspected suboptimally treated coeliac disease. Gut 1995; 37: 220 4.
  • 87
    Pistorius LR, Sweidan WH, Purdie DW, et al. Coeliac disease and bone mineral density in adult female patients. Gut 1995; 37: 639 42.
  • 88
    Walters JRF. Bone mineral density in coeliac disease. Gut 1994; 35: 150 1.
  • 89
    McFarlane XA, Bhalla AK, Reeves DE, Morgan LM, Robertson DA. Osteoporosis in treated adult coeliac disease. Gut 1995; 36: 710 14.
  • 90
    Kanis JA, Melton LJ, Christiansen C, Johnston CC, Khaltaev N. The diagnosis of osteoporosis. J Bone Miner Res 1994; 9: 1137 41.
  • 91
    Lindh E, Ljunghall S, Larsson K, Lavo B. Screening for antibodies against gliadin in patients with osteoporosis. J Intern Med 1992; 231: 403 6.
  • 92
    Molteni N, Caraceni MP, Bardella MT, Ortolani S, Gandolini GG, Bianchi P. Bone mineral density in adult celiac patients and the effect of gluten-free diet from childhood. Am J Gastroenterol 1990; 85: 51 3.
  • 93
    Caraceni MP, Molteni N, Bardella MT, Ortolani S, Nogara A, Bianchi PA. Bone and mineral metabolism in adult celiac disease. Am J Gastroenterol 1988; 83: 274 7.
  • 94
    Valdimarsson T, Löfman O, Toss G, Ström M. Reversal of osteopenia with diet in adult coeliac disease. Gut 1996; 38: 322 7.
  • 95
    McFarlane XA, Bhalla AK, Robertson DA. Effect of a gluten free diet on osteopenia in adults with newly diagnosed coeliac disease. Gut 1996; 39: 180 4.
  • 96
    Mautalen C, Gonzalez D, Mazure R, et al. Effect of treatment on bone mass, mineral metabolism, and body composition in untreated celiac disease patients. Am J Gastroenterol 1997; 92: 313 18.
  • 97
    Bai JC, Gonzalez D, Mautalen C, et al. Long-term effect of gluten restriction on bone mineral density of patients with coeliac disease. Aliment Pharmacol Ther 1997; 11: 157 64.
  • 98
    Valdimarsson T, Toss G, Ross I, Löfman O, Ström M. Bone mineral density in coeliac disease. Scand J Gastroenterol 1994; 29: 457 61.
  • 99
    Mora S, Barera G, Ricotti A, Weber G, Bianchi C, Chiumello G. Reversal of low bone density with a gluten-free diet in children and adolescents with celiac disease. Am J Clin Nutr 1998; 67: 477 81.
  • 100
    Marsh MN. Bone disease and gluten sensitivity: time to act, to treat, and to prevent. Am J Gastroenterol 1994; 89: 2105 7.
  • 101
    Riggs BL & Melton LJ III. The prevention and treatment of osteoporosis. N Engl J Med 1992; 327: 620 7.
  • 102
    Eastell R. Treatment of postmenopausal osteoporosis. N Engl J Med 1998; 338: 736 46.
  • 103
    Corazza GR, Amorosa L, Strocchi A, et al. Assessment of prevalence, familial occurrence, and possible factors inducing splenic hypofunction in adult coeliac disease in Italy. Gut 1989; 30: A1452 3(Abstract).
  • 104
    O’Grady JG, Stevens FM, Harding B, et al. Hyposplenism and gluten-sensitive enteropathy. Natural history, incidence, and relationship to diet and small bowel morphology. Gastroenterology 1984; 87: 1325 31.
  • 105
    Ferguson A, Hutton MM, Maxwell JD, Murray D. Adult coeliac disease in hyposplenic patients. Lancet 1970; i: 163–4.
  • 106
    Corazza GR, Frisoni M, Vaira D, Gasbarrini G. Effect of gluten-free diet on splenic hypofunction of adult coeliac disease. Gut 1983; 24: 228 30.
  • 107
    Corazza GR, Biagi F, Volta U, et al. Non-malignant complications of coeliac disease. In: Mäki M, Collin P, Visakorpi JK, eds. Coeliac Disease. Tampere: Coeliac Disease Study Group, 1997: 95–6.
  • 108
    Mayberry JF. Managing coeliac disease. Int J Gastroenterol 1997; 2: 9 11.