- Email: [email protected]
Autoimmune enteropathy and villous atrophy in adults
THE LANCET
Early report
Autoimmune enteropathy and villous atrophy in adults
Gino R Corazza, Federico Biagi, Umberto Volta, Maria L Andreani, Lucia De Franceschi, Giovanni Gasbarrini
Summary Background Autoimmune enteropathy is a condition described in children and characterised by villous atrophy, which is unresponsive to any dietary restrictions, and by the presence of enterocyte autoantibodies. We report two adult patients who fulfilled all the criteria for the diagnosis of this disorder. Methods Over the past 5 years we have seen four adult patients (all women, median age 51·5 [range 38–64] years) with subtotal villous atrophy, which was unresponsive to a gluten-free diet. The patients were HLADQ2 positive. IgA antigliadin and antiendomysial antibodies were not found in any of the patients. We did an indirect immunofluorescence search for enterocyte autoantibodies on monkey jejunum and for other autoantibodies for all four patients. Findings Of the four patients, two were positive for enterocyte autoantibodies and one of these two patients was positive for antiactin, antiparietal cell, and antithyroid microsomal autoantibodies. Interpretation To the best of our knowledge the two patients affected by severe enteropathy, who had never responded to any exclusion diet, and who were positive for enterocyte autoantibodies are the first cases of autoimmune enteropathy described in adults. We propose that adult patients whose disorders are unresponsive to a gluten-free diet should be tested for enterocyte autoantibodies.
Lancet 1997; 350: 106–09
Introduction The key to the diagnosis of coeliac disease is the unequivocal demonstration of histological and clinical improvement after the patient is put on a gluten-free diet.1 However, in a minority of adult patients jejunal villous atrophy and malabsorption symptoms are unresolved despite strict adherence to a gluten-free diet.2 Since most of these patients are resistant to dietary therapy from its inception, unclassified coeliac disease is the preferred term for this disorder.1,3 Here we describe four adult patients with small-bowel villous atrophy attributed to unclassified coeliac disease. Department of Internal Medicine, University of L’Aquila, via S Sisto 22E, 67100 L’Aquila, Italy (Prof G R Corazza MD, F Biagi MD, M L Andreani MD); Department of Internal Medicine, University of Bologna, Bologna (U Volta MD, L De Franceschi MD); and the Department of Internal Medicine, Catholic University of Rome, Rome (G Gasbarini MD) Correspondence to: Prof Gino R Corazza
106
Methods Case reports During the past 5 years, we have seen four women with features of unclassified coeliac disease (table 1). They all proved to be HLA-DQ2 positive. Intestinal biopsy samples were taken because of severe malabsorption symptoms. Three of them (patients one, two, and three) were referred to us because of a lack of histological and clinical response to gluten withdrawal from the diet and were therefore already on gluten-free diet at the time of our first observation. Adherence to the gluten-free diet was carefully monitored and judged to be good in all patients. Hypersensitivity to other foods, anecdotally reported as being capable of delaying intestinal mucosa healing, was excluded. Small-bowel barium enema and abdominal computedtomography scan excluded the presence of ulcerative jejunoileitis or intestinal lymphoma. Table 1 gives the surface area/volume ratio of the jejunal mucosa relative to the results of the first intestinal biopsy sample we took. Both the surface area/volume ratio, which is a measure of the degree of villous atrophy and crypt hypertrophy,4 and the percentage of intraepithelial lymphocytes5 were abnormal for all four patients. For patient four, who had her first biopsy sample taken whilst she was on a gluten-containing diet, gluten withdrawal did not lead to any improvement in the symptoms and a further biopsy sample taken after 16 months of diet showed lesions similar to those seen on the first biopsy sample. A gluten-free diet was combined with 3-methylprednisolone at the initial daily dose of 50 mg in patient one and 40 mg in patients two, three, and four.6 This regimen led to a substantial improvement in clinical symptoms, permitting a gradual reduction, at different rates in the four patients, of the steroid therapy to a maintenance dose of 12 mg daily of deflazacort (Hoechst Marion Roussel, Uxbridge, UK). In patients one and two, the overall steroid requirement was greater than in patients three and four. After at least 1 year of steroid treatment, intestinal biopsy samples showed an improvement in jejunal morphology in all four patients. Patient one died from posterior myocardial infarct due to the thrombotic occlusion of the posterolateral branch of the circumflex coronary artery after 18 months of steroid therapy. Patients two, three, and four who were on a gluten-free diet and deflazacort were in a good clinical condition at the end of the study. Patients one and two had a distinctive serological pattern that proved essential for their diagnosis.
Enterocyte autoantibodies We tested for enterocyte autoantibodies by indirect immunofluorescence on 5 m cryostat sections of monkey jejunum (Medic, Turin, Italy). Sections were incubated with serum samples at the initial dilution of one in five for 30 min at room temperature, then washed three times for 5 min in phosphate-buffered saline and stained separately with fluorescein isothiocyanate conjugated (FITC) rabbit antibodies to human IgA, IgG, and IgM (Dako, Copenhagen, Denmark). The appropriate working dilution of FITC antibodies to human IgA, IgG, and IgM was 1 to 100, because a more concentrated conjugate produced a non-specific fluorescence on primate tissue. A single-layer FITC was used to detect the presence of cross-reactivity between antibodies physiologically present in
Vol 350 • July 12th, 1997
THE LANCET
Patient one
Patient two
Patient three
Patient four
Age (years)
38
47
56
64
HLA status
DQ2
DQ2
DQ2
DQ2
Months of gluten-free diet
23
25
18
··
Jejunal mucosa surface area/volume ratio* Presteroid† 36·9 Poststeroid† 20·4
38·5 37·5
11·0 37·3
10·0 45·5
% jejunal intraepithelial lymphocytes‡ Presteroid† 46 Poststeroid† 37
42 29
51 22
45 25
*Abnormal <17·7. †Steroid=3-methylprednisolone. ‡Abnormal >30%.
Table 1: Clinical and histological features of four women with unclassified coeliac disease primate tissue, and FITC antihuman immunoglobulins. After further washing (three times for 5 min in phosphate-buffered saline), the slides were mounted in phosphate-buffered saline (one part) and glycerin buffer (one part) and observed under a Leitz microscope with vertical illumination (Xenon XBO 75-W lamp). Positive sera were titrated up to the endpoint. For the detection of complement-fixing enterocyte autoantibodies undiluted serum samples from patients were applied to monkey jejunum sections for 30 min, followed by fresh normal serum as a source of additional complement. The fresh normal serum was obtained from a pool of sera taken from laboratory members previously shown to have no enterocyte autoantibodies. FITC rabbit antibodies to human C3
Antibodies IgA antigliadin IgG antigliadin IgA antiendomysium Autoantibodies IgA-enterocyte Presteroid Poststeroid* IgG enterocyte Presteroid Poststeroid Other Presteroid
Patient one
Patient two
Patient three
Patient four
⫺ ⫺ ⫺
⫺ ⫺ ⫺
⫺ + ⫺
⫺ + ⫺
1/20 1/20
1/320 1/80
⫺ ⫺
⫺ ⫺
1/40 1/40
1/640 1/40
⫺ ⫺
⫺ ⫺
⫺
⫺
⫺
⫺
Poststeroid*
Antiactin 1/320 Antiparietal cell 1/8 Antithyroid microsomal 1/16 Antiactin 1/80 Antiparietal cell 1/8 Antithyroid microsomal 1/16
*Steroid=3-methylprednisolone.
Table 2: Serological pattern of four women with unclassified coeliac disease
(Dako) at a dilution of 1 in 20 in phosphate-buffered saline was then applied to all sections. Enterocyte autoantibodies were also tested for in the serum of 95 untreated adult coeliac patients who subsequently showed a good histological and clinical response to a gluten-free diet and in 50 healthy controls.
Other autoantibodies Organ-specific autoantibodies, including antigastric parietal cell, thyroid microsomal, pancreatic islet cell, and adrenal cortex antibodies, were detected by indirect immunofluorescence. We tested serum samples diluted one in two in phosphate-buffered saline on sections of human stomach, thyroid, pancreas, and adrenal cortex tissue obtained from patients with blood group O, taken during surgical procedures. We tested for non-organ-specific autoantibodies, including antinuclear, smooth muscle, mitochondrial, liver-kidney microsomal, and ribosomal antibodies, by indirect immunofluorescence on rodent liver, kidney, and stomach with serum samples diluted one in ten in phospate-buffered saline. Positive sera were titrated up to the endpoint.
Antigliadin and antiendomysial antibodies IgA and IgG serum antigliadin antibodies were measured by a micro-ELISA test as previously described.7 IgA antiendomysial antibodies were detected by indirect immunofluorescence with commercially available cryostat sections of monkey oesophagus as the antigen (BioSystems, Milan, Italy).8 Serum containing antibody at a titre of one in five or more was taken as positive. In our laboratory the antigliadin and antiendomysial antibodies sensitivity and specificity values, previously verified in a large panel of patients, are 92% and 90% for antigliadin and 99% and 100% for antiendomysial antibodies.9
Results
Figure 1: Immunofluorescent pattern of IgA enterocyte autoantibody in patient two Widespread staining of enterocyte cytoplasm with a stronger positivity at brush-border level is evident on cryostat section of monkey jejunum (25⫻).
Vol 350 • July 12th, 1997
The serological pattern of our patients is shown in table 2. Patients one and two were positive for the presence of enterocyte autoantibodies. Only patients three and four were positive for IgG antigliadin antibodies, but IgA antigliadin and antiendomysial antibodies were absent in all four patients. Enterocyte autoantibodies belonged to IgA and IgG classes, whereas the test for IgM enterocyte autoantibodies was always negative. Only patient two, who had the highest enterocyte autoantibody titres, showed the ability to fix the complement and proved to be positive also for the presence of other organ-specific and non-organ-specific autoantibodies. 107
THE LANCET
Figure 2: Immunofluorescence pattern of villous epithelium of monkey jejunum with a negative control serum (25⫻)
Figure 1 shows the immunofluorescent pattern of IgA enterocyte autoantibody in patient two; figure 2 shows the absence of fluorescence with a negative control serum. Figure 3 shows the IgA class enterocyte autoantibody in patient one. Serum from patients three and four, from 95 untreated uncomplicated patients with coeliac disease, and from 50 healthy controls turned out to be negative when tested for enterocyte autoantibodies. In patients one and two, autoantibody titres were reassessed (table 2) more than a year after the addition of steroid therapy to a gluten-free diet. With this regimen, which had led to a clinical and histological improvement in both patients, enterocyte-autoantibody titres showed no variations in patient one, whereas a substantial drop in enterocyte autoantibodies and antiactin autoantibody titres was evident in patient two.
Discussion We report four adult patients with villous atrophy initially diagnosed as unclassified coeliac disease. It was subsequently clear that two of these patients (one and two) fulfilled all the criteria proposed by Unsworth and Walker-Smith10 for the diagnosis of autoimmune enteropathy. In patient two predisposition to autoimmune disease, as well as the presence of circulating enterocyte autoantibodies, was also indicated by the presence of other organ-specific and non-organ-specific autoantibodies, but clinical and laboratory investigations did 108
Figure 3: Immunofluorescent pattern of IgA enterocyte autoantibody in patient one Bright staining of the brush border and a mild positivity of enterocyte cytoplasm are evident on cryostat section of monkey jejunum (25⫻).
not show the presence of other autoimmune diseases. However, it is generally recognised that autoantibodies can appear many years before the disease onset.11 Although circulating enterocyte autoantibodies have been detected in adults who do not have intestinal lesions,12 all cases of autoimmune enteropathy described to date have occurred in children,13–17 and in all of them the main feature of the jejunal biopsy samples was crypt hyperplastic villous atrophy unresponsive to a gluten-free diet. For the two patients in our study found to have enterocyte autoantibodies, the adult onset of the enteropathy was confirmed by the fact that they had never had malabsorption symptoms during their childhood. The difference between their condition and coeliac disease, in addition to the lack of response to a gluten-free diet, was underlined by the constant negative testing to antigliadin and antiendomysial antibodies. Both the patients were HLA-DQ2 positive, as are nearly all patients with coeliac disease,1 but it is known that DQ2 alleles contribute to susceptibility to many autoimmune diseases.18 Although only very cautious conclusions can be drawn on the basis of two cases, our results confirm that high enterocyte-autoantibody titres and complement-fixing ability do not correlate with the histological severity of the enteropathy.19 In both patients steroid treatment led to an improvement of the histological lesions, but only in
Vol 350 • July 12th, 1997
THE LANCET
patient two, who had the highest initial titres of enterocyte autoantibodies, was this accompanied by a reduction in autoantibody titres. The pathogenic importance of enterocyte autoantibodies remains uncertain. Enterocyte autoantibodies may appear late during the disease14,15 and therefore may be a secondary phenomenon. Since villous atrophy results from an in-situ activation of T cells,20 it has been suggested that autoreactive T-cell clones also play a major part in the generation of the intestinal lesions of autoimmune enteropathy. Supporting this theory, features of intestinal T-cell activation, such as inappropriate HLA class II molecule expression in the crypt enterocytes21,22 and the appearance of interleukin-2 receptor-bearing cells in the lamina propria,22 have been found in children with autoimmune enteropathy. Moreover, the frequent detection of enterocyte autoantibody in adults with HIV infection23 raises the possibility that this disorder reflects a constitutive or acquired disturbance of extrathymic T-cell differentiation, localised to the intestine. The increased percentage of intraepithelial lymphocytes found in our two patients stresses the pathogenic importance of cellmediated mechanisms. In the paediatric form of autoimmune enteropathy two studies have dealt with the number of CD3-positive intraepithelial lymphocytes, which was found to be not increased19 or clearly increased.22 Both these studies, however, agree that ␥␦positive intraepithelial lymphocytes are not increased in this condition and that this finding can help to distinguish it from coeliac disease. We must stress that patients three and four, who were negative when tested for enterocyte autoantibodies, did not differ from the two patients with autoimmune enteropathy either in terms of severity of the histological lesions or clinical course. Since other known causes of flat jejunal mucosa24 had been excluded, the possibility remained that these patients had autoimmune enteropathy, in which circulating enterocyte autoantibodies have not yet appeared14,15 or, more probably, had coeliac disease that had already become unresponsive to a gluten-free diet before diagnosis. In the absence of other evidence we believe that it is correct to maintain the diagnosis of unclassified coeliac disease for patients three and four. As far as we know, patients one and two show for the first time that the development of an autoimmune enteropathy can occur in adulthood. We believe that systematic testing for enterocyte autoantibodies in adult patients with flat mucosa unresponsive to a gluten-free diet will lead to the diagnosis of other patients with the adult form of autoimmune enteropathy.
References 1 2 3 4
5 6
7
8
9
10 11
12
13
14
15
16
17
18 19
20
21
Contributors Prof G R Corazza was the principal investigator and designed the study; he wrote the paper with Prof G Gasbarrini. F Biagi and M L Andreani followed the patients clinically, carried out intestinal biopsies and mucosal morphometry, and tested antiendomysial and antigliadin antibodies. U Volta and L De Franceschi did enterocyte autoantibody immunofluorescence assays.
22
23
Acknowledgments We thank J A Walker-Smith (University Department of Paediatric Gastroenterology, Royal Free Hospital, London) for help and advice.
Vol 350 • July 12th, 1997
24
Trier JS. Celiac sprue. N Engl J Med 1991; 325: 1709–29. Trier JS, Falchuk ZM, Carey MC, Schreiber DS. Celiac sprue and refractory sprue. Gastroenterology 1978; 75: 307–16. Rubin CE. Sprue by any other name. Gastroenterology 1970; 58: 409–13. Corazza GR, Frazzoni M, Dixon MF, Gasbarrini G. Quantitative assessment of the mucosal architecture of jejunal biopsy specimens: a comparison between linear measurement, stereology, and computer aided microscopy. J Clin Pathol 1985; 38: 765–70. Ferguson A. Intraepithelial lymphocytes of the small intestine. Gut 1977; 18: 921–37. 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. Volta U, Lenzi M, Lazzari R, et al. Antibodies to gliadin detected by immunofluorescence and a micro-ELISA method: markers of active childhood and adult coeliac disease. Gut 1985; 26: 667–71. Chorzelski TP, Beutner EH, Sulej J, et al. IgA-antiendomysium antibody. A new immunological marker of dermatitis herpetiformis and coeliac disease. Br J Dermatol 1984; 111: 395–402. Corazza GR, Valentini RA, Andreani ML, et al. Subclinical coeliac disease is a frequent cause of iron-deficiency anaemia. Scand J Gastroenterol 1995; 30: 153–56. Unsworth DJ, Walker-Smith JA. Autoimmunity in diarrhoeal disease. J Pediatr Gastroenterol Nutr 1985; 4: 375–80. Bottazzo GF, Todd I, Mirakian R, Belfiore A, Pujol-Borrell R. Organspecific autoimmunity: a 1986 overview. Immunol Rev 1986; 94: 137–69. Martin-Villa JM, Regueiro JR, De Juan D, et al. T-lymphocyte dysfunctions occurring together with apical gut epithelial cell autoantibodies. Gastroenterology 1991; 101: 390–97. McCarthy DM, Katz SI, Gazze L, Waldmann TA, Nelson DL, Strober W. Selective IgA deficiency associated with total villous atrophy of the small intestine and organ-specific anti-epithelial cell antibody. J Immunol 1978; 120: 932–38. Walker-Smith JA, Unsworth DJ, Hutchins P, Phillips AD, Holborow EJ. Autoantibodies against gut epithelium in a child with small intestinal enteropathy. Lancet 1982; i: 566–67. Savilahti E, Pelkonen P, Holmberg C, Perkkiö M, Unsworth J. Fatal unresponsive villous atrophy of the jejunum, connective tissue disease and diabetes in a girl with intestinal epithelial cell antibody. Acta Paediatr Scand 1985; 74: 472–76. Mirakian R, Richardson A, Milla PJ, et al. Protracted diarrhoea of infancy: evidence in support of an autoimmune variant. BMJ 1986; 293: 1132–36. Catassi C, Mirakian R, Natalini G, et al. Unresponsive enteropathy associated with circulating enterocyte autoantibodies in a boy with common variable hypogammaglobulinemia and type I diabetes. J Pediatr Gastroenterol Nutr 1988; 7: 608–13. Nepom GT, Erlich H. MHC Class-II molecules and autoimmunity. Annu Rev Immunol 1991; 9: 493–525. Walker-Smith JA. Coeliac disease and autoimmune enteropathy. In: Mearin ML, Mulder CJJ, eds. Coeliac disease. Leiden: Kluwer Academic Publishers, 1991: 131–36. MacDonald TT, Spencer J. Evidence that activated mucosal T cells play a role in the pathogenesis of enteropathy in human small intestine. J Exp Med 1988; 167: 1341–49. Mirakian R, Hill S, Richardson A, Milla PJ, Walker-Smith JA, Bottazzo GF. HLA product expression and lymphocyte subpopulations in jejunum biopsies of children with idiopathic protracted diarrhoea and enterocyte autoantibodies. J Autoimmun 1988; 1: 263–77. Cuenod B, Brousse N, Goulet O, et al. Classification of intractable diarrhea in infancy using clinical and immunohistological criteria. Gastroenterology 1990; 99: 1037–43. Martin-Villa JM, Camblor S, Costa R, Arnaiz-Villena A. Gut epithelial cell autoantibodies in AIDS pathogenesis. Lancet 1993; 342: 380. Katz AJ, Grand RJ. All that flattens is not “sprue”. Gastroenterology 1979; 76: 375–77.
109
Early report
Autoimmune enteropathy and villous atrophy in adults
Gino R Corazza, Federico Biagi, Umberto Volta, Maria L Andreani, Lucia De Franceschi, Giovanni Gasbarrini
Summary Background Autoimmune enteropathy is a condition described in children and characterised by villous atrophy, which is unresponsive to any dietary restrictions, and by the presence of enterocyte autoantibodies. We report two adult patients who fulfilled all the criteria for the diagnosis of this disorder. Methods Over the past 5 years we have seen four adult patients (all women, median age 51·5 [range 38–64] years) with subtotal villous atrophy, which was unresponsive to a gluten-free diet. The patients were HLADQ2 positive. IgA antigliadin and antiendomysial antibodies were not found in any of the patients. We did an indirect immunofluorescence search for enterocyte autoantibodies on monkey jejunum and for other autoantibodies for all four patients. Findings Of the four patients, two were positive for enterocyte autoantibodies and one of these two patients was positive for antiactin, antiparietal cell, and antithyroid microsomal autoantibodies. Interpretation To the best of our knowledge the two patients affected by severe enteropathy, who had never responded to any exclusion diet, and who were positive for enterocyte autoantibodies are the first cases of autoimmune enteropathy described in adults. We propose that adult patients whose disorders are unresponsive to a gluten-free diet should be tested for enterocyte autoantibodies.
Lancet 1997; 350: 106–09
Introduction The key to the diagnosis of coeliac disease is the unequivocal demonstration of histological and clinical improvement after the patient is put on a gluten-free diet.1 However, in a minority of adult patients jejunal villous atrophy and malabsorption symptoms are unresolved despite strict adherence to a gluten-free diet.2 Since most of these patients are resistant to dietary therapy from its inception, unclassified coeliac disease is the preferred term for this disorder.1,3 Here we describe four adult patients with small-bowel villous atrophy attributed to unclassified coeliac disease. Department of Internal Medicine, University of L’Aquila, via S Sisto 22E, 67100 L’Aquila, Italy (Prof G R Corazza MD, F Biagi MD, M L Andreani MD); Department of Internal Medicine, University of Bologna, Bologna (U Volta MD, L De Franceschi MD); and the Department of Internal Medicine, Catholic University of Rome, Rome (G Gasbarini MD) Correspondence to: Prof Gino R Corazza
106
Methods Case reports During the past 5 years, we have seen four women with features of unclassified coeliac disease (table 1). They all proved to be HLA-DQ2 positive. Intestinal biopsy samples were taken because of severe malabsorption symptoms. Three of them (patients one, two, and three) were referred to us because of a lack of histological and clinical response to gluten withdrawal from the diet and were therefore already on gluten-free diet at the time of our first observation. Adherence to the gluten-free diet was carefully monitored and judged to be good in all patients. Hypersensitivity to other foods, anecdotally reported as being capable of delaying intestinal mucosa healing, was excluded. Small-bowel barium enema and abdominal computedtomography scan excluded the presence of ulcerative jejunoileitis or intestinal lymphoma. Table 1 gives the surface area/volume ratio of the jejunal mucosa relative to the results of the first intestinal biopsy sample we took. Both the surface area/volume ratio, which is a measure of the degree of villous atrophy and crypt hypertrophy,4 and the percentage of intraepithelial lymphocytes5 were abnormal for all four patients. For patient four, who had her first biopsy sample taken whilst she was on a gluten-containing diet, gluten withdrawal did not lead to any improvement in the symptoms and a further biopsy sample taken after 16 months of diet showed lesions similar to those seen on the first biopsy sample. A gluten-free diet was combined with 3-methylprednisolone at the initial daily dose of 50 mg in patient one and 40 mg in patients two, three, and four.6 This regimen led to a substantial improvement in clinical symptoms, permitting a gradual reduction, at different rates in the four patients, of the steroid therapy to a maintenance dose of 12 mg daily of deflazacort (Hoechst Marion Roussel, Uxbridge, UK). In patients one and two, the overall steroid requirement was greater than in patients three and four. After at least 1 year of steroid treatment, intestinal biopsy samples showed an improvement in jejunal morphology in all four patients. Patient one died from posterior myocardial infarct due to the thrombotic occlusion of the posterolateral branch of the circumflex coronary artery after 18 months of steroid therapy. Patients two, three, and four who were on a gluten-free diet and deflazacort were in a good clinical condition at the end of the study. Patients one and two had a distinctive serological pattern that proved essential for their diagnosis.
Enterocyte autoantibodies We tested for enterocyte autoantibodies by indirect immunofluorescence on 5 m cryostat sections of monkey jejunum (Medic, Turin, Italy). Sections were incubated with serum samples at the initial dilution of one in five for 30 min at room temperature, then washed three times for 5 min in phosphate-buffered saline and stained separately with fluorescein isothiocyanate conjugated (FITC) rabbit antibodies to human IgA, IgG, and IgM (Dako, Copenhagen, Denmark). The appropriate working dilution of FITC antibodies to human IgA, IgG, and IgM was 1 to 100, because a more concentrated conjugate produced a non-specific fluorescence on primate tissue. A single-layer FITC was used to detect the presence of cross-reactivity between antibodies physiologically present in
Vol 350 • July 12th, 1997
THE LANCET
Patient one
Patient two
Patient three
Patient four
Age (years)
38
47
56
64
HLA status
DQ2
DQ2
DQ2
DQ2
Months of gluten-free diet
23
25
18
··
Jejunal mucosa surface area/volume ratio* Presteroid† 36·9 Poststeroid† 20·4
38·5 37·5
11·0 37·3
10·0 45·5
% jejunal intraepithelial lymphocytes‡ Presteroid† 46 Poststeroid† 37
42 29
51 22
45 25
*Abnormal <17·7. †Steroid=3-methylprednisolone. ‡Abnormal >30%.
Table 1: Clinical and histological features of four women with unclassified coeliac disease primate tissue, and FITC antihuman immunoglobulins. After further washing (three times for 5 min in phosphate-buffered saline), the slides were mounted in phosphate-buffered saline (one part) and glycerin buffer (one part) and observed under a Leitz microscope with vertical illumination (Xenon XBO 75-W lamp). Positive sera were titrated up to the endpoint. For the detection of complement-fixing enterocyte autoantibodies undiluted serum samples from patients were applied to monkey jejunum sections for 30 min, followed by fresh normal serum as a source of additional complement. The fresh normal serum was obtained from a pool of sera taken from laboratory members previously shown to have no enterocyte autoantibodies. FITC rabbit antibodies to human C3
Antibodies IgA antigliadin IgG antigliadin IgA antiendomysium Autoantibodies IgA-enterocyte Presteroid Poststeroid* IgG enterocyte Presteroid Poststeroid Other Presteroid
Patient one
Patient two
Patient three
Patient four
⫺ ⫺ ⫺
⫺ ⫺ ⫺
⫺ + ⫺
⫺ + ⫺
1/20 1/20
1/320 1/80
⫺ ⫺
⫺ ⫺
1/40 1/40
1/640 1/40
⫺ ⫺
⫺ ⫺
⫺
⫺
⫺
⫺
Poststeroid*
Antiactin 1/320 Antiparietal cell 1/8 Antithyroid microsomal 1/16 Antiactin 1/80 Antiparietal cell 1/8 Antithyroid microsomal 1/16
*Steroid=3-methylprednisolone.
Table 2: Serological pattern of four women with unclassified coeliac disease
(Dako) at a dilution of 1 in 20 in phosphate-buffered saline was then applied to all sections. Enterocyte autoantibodies were also tested for in the serum of 95 untreated adult coeliac patients who subsequently showed a good histological and clinical response to a gluten-free diet and in 50 healthy controls.
Other autoantibodies Organ-specific autoantibodies, including antigastric parietal cell, thyroid microsomal, pancreatic islet cell, and adrenal cortex antibodies, were detected by indirect immunofluorescence. We tested serum samples diluted one in two in phosphate-buffered saline on sections of human stomach, thyroid, pancreas, and adrenal cortex tissue obtained from patients with blood group O, taken during surgical procedures. We tested for non-organ-specific autoantibodies, including antinuclear, smooth muscle, mitochondrial, liver-kidney microsomal, and ribosomal antibodies, by indirect immunofluorescence on rodent liver, kidney, and stomach with serum samples diluted one in ten in phospate-buffered saline. Positive sera were titrated up to the endpoint.
Antigliadin and antiendomysial antibodies IgA and IgG serum antigliadin antibodies were measured by a micro-ELISA test as previously described.7 IgA antiendomysial antibodies were detected by indirect immunofluorescence with commercially available cryostat sections of monkey oesophagus as the antigen (BioSystems, Milan, Italy).8 Serum containing antibody at a titre of one in five or more was taken as positive. In our laboratory the antigliadin and antiendomysial antibodies sensitivity and specificity values, previously verified in a large panel of patients, are 92% and 90% for antigliadin and 99% and 100% for antiendomysial antibodies.9
Results
Figure 1: Immunofluorescent pattern of IgA enterocyte autoantibody in patient two Widespread staining of enterocyte cytoplasm with a stronger positivity at brush-border level is evident on cryostat section of monkey jejunum (25⫻).
Vol 350 • July 12th, 1997
The serological pattern of our patients is shown in table 2. Patients one and two were positive for the presence of enterocyte autoantibodies. Only patients three and four were positive for IgG antigliadin antibodies, but IgA antigliadin and antiendomysial antibodies were absent in all four patients. Enterocyte autoantibodies belonged to IgA and IgG classes, whereas the test for IgM enterocyte autoantibodies was always negative. Only patient two, who had the highest enterocyte autoantibody titres, showed the ability to fix the complement and proved to be positive also for the presence of other organ-specific and non-organ-specific autoantibodies. 107
THE LANCET
Figure 2: Immunofluorescence pattern of villous epithelium of monkey jejunum with a negative control serum (25⫻)
Figure 1 shows the immunofluorescent pattern of IgA enterocyte autoantibody in patient two; figure 2 shows the absence of fluorescence with a negative control serum. Figure 3 shows the IgA class enterocyte autoantibody in patient one. Serum from patients three and four, from 95 untreated uncomplicated patients with coeliac disease, and from 50 healthy controls turned out to be negative when tested for enterocyte autoantibodies. In patients one and two, autoantibody titres were reassessed (table 2) more than a year after the addition of steroid therapy to a gluten-free diet. With this regimen, which had led to a clinical and histological improvement in both patients, enterocyte-autoantibody titres showed no variations in patient one, whereas a substantial drop in enterocyte autoantibodies and antiactin autoantibody titres was evident in patient two.
Discussion We report four adult patients with villous atrophy initially diagnosed as unclassified coeliac disease. It was subsequently clear that two of these patients (one and two) fulfilled all the criteria proposed by Unsworth and Walker-Smith10 for the diagnosis of autoimmune enteropathy. In patient two predisposition to autoimmune disease, as well as the presence of circulating enterocyte autoantibodies, was also indicated by the presence of other organ-specific and non-organ-specific autoantibodies, but clinical and laboratory investigations did 108
Figure 3: Immunofluorescent pattern of IgA enterocyte autoantibody in patient one Bright staining of the brush border and a mild positivity of enterocyte cytoplasm are evident on cryostat section of monkey jejunum (25⫻).
not show the presence of other autoimmune diseases. However, it is generally recognised that autoantibodies can appear many years before the disease onset.11 Although circulating enterocyte autoantibodies have been detected in adults who do not have intestinal lesions,12 all cases of autoimmune enteropathy described to date have occurred in children,13–17 and in all of them the main feature of the jejunal biopsy samples was crypt hyperplastic villous atrophy unresponsive to a gluten-free diet. For the two patients in our study found to have enterocyte autoantibodies, the adult onset of the enteropathy was confirmed by the fact that they had never had malabsorption symptoms during their childhood. The difference between their condition and coeliac disease, in addition to the lack of response to a gluten-free diet, was underlined by the constant negative testing to antigliadin and antiendomysial antibodies. Both the patients were HLA-DQ2 positive, as are nearly all patients with coeliac disease,1 but it is known that DQ2 alleles contribute to susceptibility to many autoimmune diseases.18 Although only very cautious conclusions can be drawn on the basis of two cases, our results confirm that high enterocyte-autoantibody titres and complement-fixing ability do not correlate with the histological severity of the enteropathy.19 In both patients steroid treatment led to an improvement of the histological lesions, but only in
Vol 350 • July 12th, 1997
THE LANCET
patient two, who had the highest initial titres of enterocyte autoantibodies, was this accompanied by a reduction in autoantibody titres. The pathogenic importance of enterocyte autoantibodies remains uncertain. Enterocyte autoantibodies may appear late during the disease14,15 and therefore may be a secondary phenomenon. Since villous atrophy results from an in-situ activation of T cells,20 it has been suggested that autoreactive T-cell clones also play a major part in the generation of the intestinal lesions of autoimmune enteropathy. Supporting this theory, features of intestinal T-cell activation, such as inappropriate HLA class II molecule expression in the crypt enterocytes21,22 and the appearance of interleukin-2 receptor-bearing cells in the lamina propria,22 have been found in children with autoimmune enteropathy. Moreover, the frequent detection of enterocyte autoantibody in adults with HIV infection23 raises the possibility that this disorder reflects a constitutive or acquired disturbance of extrathymic T-cell differentiation, localised to the intestine. The increased percentage of intraepithelial lymphocytes found in our two patients stresses the pathogenic importance of cellmediated mechanisms. In the paediatric form of autoimmune enteropathy two studies have dealt with the number of CD3-positive intraepithelial lymphocytes, which was found to be not increased19 or clearly increased.22 Both these studies, however, agree that ␥␦positive intraepithelial lymphocytes are not increased in this condition and that this finding can help to distinguish it from coeliac disease. We must stress that patients three and four, who were negative when tested for enterocyte autoantibodies, did not differ from the two patients with autoimmune enteropathy either in terms of severity of the histological lesions or clinical course. Since other known causes of flat jejunal mucosa24 had been excluded, the possibility remained that these patients had autoimmune enteropathy, in which circulating enterocyte autoantibodies have not yet appeared14,15 or, more probably, had coeliac disease that had already become unresponsive to a gluten-free diet before diagnosis. In the absence of other evidence we believe that it is correct to maintain the diagnosis of unclassified coeliac disease for patients three and four. As far as we know, patients one and two show for the first time that the development of an autoimmune enteropathy can occur in adulthood. We believe that systematic testing for enterocyte autoantibodies in adult patients with flat mucosa unresponsive to a gluten-free diet will lead to the diagnosis of other patients with the adult form of autoimmune enteropathy.
References 1 2 3 4
5 6
7
8
9
10 11
12
13
14
15
16
17
18 19
20
21
Contributors Prof G R Corazza was the principal investigator and designed the study; he wrote the paper with Prof G Gasbarrini. F Biagi and M L Andreani followed the patients clinically, carried out intestinal biopsies and mucosal morphometry, and tested antiendomysial and antigliadin antibodies. U Volta and L De Franceschi did enterocyte autoantibody immunofluorescence assays.
22
23
Acknowledgments We thank J A Walker-Smith (University Department of Paediatric Gastroenterology, Royal Free Hospital, London) for help and advice.
Vol 350 • July 12th, 1997
24
Trier JS. Celiac sprue. N Engl J Med 1991; 325: 1709–29. Trier JS, Falchuk ZM, Carey MC, Schreiber DS. Celiac sprue and refractory sprue. Gastroenterology 1978; 75: 307–16. Rubin CE. Sprue by any other name. Gastroenterology 1970; 58: 409–13. Corazza GR, Frazzoni M, Dixon MF, Gasbarrini G. Quantitative assessment of the mucosal architecture of jejunal biopsy specimens: a comparison between linear measurement, stereology, and computer aided microscopy. J Clin Pathol 1985; 38: 765–70. Ferguson A. Intraepithelial lymphocytes of the small intestine. Gut 1977; 18: 921–37. 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. Volta U, Lenzi M, Lazzari R, et al. Antibodies to gliadin detected by immunofluorescence and a micro-ELISA method: markers of active childhood and adult coeliac disease. Gut 1985; 26: 667–71. Chorzelski TP, Beutner EH, Sulej J, et al. IgA-antiendomysium antibody. A new immunological marker of dermatitis herpetiformis and coeliac disease. Br J Dermatol 1984; 111: 395–402. Corazza GR, Valentini RA, Andreani ML, et al. Subclinical coeliac disease is a frequent cause of iron-deficiency anaemia. Scand J Gastroenterol 1995; 30: 153–56. Unsworth DJ, Walker-Smith JA. Autoimmunity in diarrhoeal disease. J Pediatr Gastroenterol Nutr 1985; 4: 375–80. Bottazzo GF, Todd I, Mirakian R, Belfiore A, Pujol-Borrell R. Organspecific autoimmunity: a 1986 overview. Immunol Rev 1986; 94: 137–69. Martin-Villa JM, Regueiro JR, De Juan D, et al. T-lymphocyte dysfunctions occurring together with apical gut epithelial cell autoantibodies. Gastroenterology 1991; 101: 390–97. McCarthy DM, Katz SI, Gazze L, Waldmann TA, Nelson DL, Strober W. Selective IgA deficiency associated with total villous atrophy of the small intestine and organ-specific anti-epithelial cell antibody. J Immunol 1978; 120: 932–38. Walker-Smith JA, Unsworth DJ, Hutchins P, Phillips AD, Holborow EJ. Autoantibodies against gut epithelium in a child with small intestinal enteropathy. Lancet 1982; i: 566–67. Savilahti E, Pelkonen P, Holmberg C, Perkkiö M, Unsworth J. Fatal unresponsive villous atrophy of the jejunum, connective tissue disease and diabetes in a girl with intestinal epithelial cell antibody. Acta Paediatr Scand 1985; 74: 472–76. Mirakian R, Richardson A, Milla PJ, et al. Protracted diarrhoea of infancy: evidence in support of an autoimmune variant. BMJ 1986; 293: 1132–36. Catassi C, Mirakian R, Natalini G, et al. Unresponsive enteropathy associated with circulating enterocyte autoantibodies in a boy with common variable hypogammaglobulinemia and type I diabetes. J Pediatr Gastroenterol Nutr 1988; 7: 608–13. Nepom GT, Erlich H. MHC Class-II molecules and autoimmunity. Annu Rev Immunol 1991; 9: 493–525. Walker-Smith JA. Coeliac disease and autoimmune enteropathy. In: Mearin ML, Mulder CJJ, eds. Coeliac disease. Leiden: Kluwer Academic Publishers, 1991: 131–36. MacDonald TT, Spencer J. Evidence that activated mucosal T cells play a role in the pathogenesis of enteropathy in human small intestine. J Exp Med 1988; 167: 1341–49. Mirakian R, Hill S, Richardson A, Milla PJ, Walker-Smith JA, Bottazzo GF. HLA product expression and lymphocyte subpopulations in jejunum biopsies of children with idiopathic protracted diarrhoea and enterocyte autoantibodies. J Autoimmun 1988; 1: 263–77. Cuenod B, Brousse N, Goulet O, et al. Classification of intractable diarrhea in infancy using clinical and immunohistological criteria. Gastroenterology 1990; 99: 1037–43. Martin-Villa JM, Camblor S, Costa R, Arnaiz-Villena A. Gut epithelial cell autoantibodies in AIDS pathogenesis. Lancet 1993; 342: 380. Katz AJ, Grand RJ. All that flattens is not “sprue”. Gastroenterology 1979; 76: 375–77.
109