1487

Najlerahim A, Bowen DM. Biochemical measurements in Alzheimer’s disease reveal a necessity for improved neurolmaging techniques to study metabolism. Biochem J 1988; 251: 305-08. 4. Duara R, Grady C, Haxby J, et al. Positron emission tomography in Alzheimer’s disease. Neurology 1986; 36: 879-87. 5. Foster NL, Chase TN, Mansi L, et al. Cortical abnormalities in Alzheimer’s disease. Ann Neurol 1984; 16: 649-54. 6. Friedland RP, Budinger TF, Ganz E, et al. Regional cerebral metabolic alterations in dementia of the Alzheimer type: positron emission tomography with (18F) fluorodeoxyglucose. J Comput Assist Tomogr 1983; 7: 590-98. 7. Frackowiak RSJ, Pozzilli C, Legg NJ, et al. Regional cerebral oxygen supply and utilization in dementia. Brain 1981; 104: 753-78. 8. Neirinckx RD, Canning LR, Piper IM, et al. Technetium-99m d, 1-HM-PAO: a new radiopharmaceutical for SPECT imaging of regional cerebral blood perfusion. J Nucl Med 1987; 28: 191-202. 9. Burns A, Philpot MP, Costa DC, Ell PJ, Levy R. The investigation of Alzheimer’s disease with single photon emission tomography. J Neurol Neurosurg Psychiatry 1989; 52: 248-53. 10. Hunter R, McLuskie R, Wyper D, et al. The pattern of function-related regional cerebral blood flow investigated by single photon emission tomography with 99mTc-HMPAO in patients with presenile Alzheimer’s disease and Korsakoff’s psychosis. Psychol Med 1989; 19: 3.

847-55.

11. Neary D, Snowden JS, Shields RA, et al. Single photon emission tomography using 99mTc-HM-PAO in the investigation of dementia. J Neurol Neurosurg Psychiatry 1987; 50: 1101-09.

12. American Psychiatric Association. Diagnostic and statistical manual of mental disorders, 3rd revised. Washington DC: American Psychiatric Association, 1987. 13. McKhann G, Drachman D, Folstein M, et al. Clinical diagnosis of Alzheimer’s disease: report of the NINCDS-ADRDA Workgroup under the auspices of Department of Health and Human Services Task Force on Alzheimer’s Disease. Neurology 1984; 34: 939-44. 14. Hachinski VC, Iliff LD, Zilkha E, et al. Cerebral blood flow in dementia. Arch Neurol 1975; 32: 632-37. 15. Roth M, Tym E, Mountjoy CQ, et al. CAMDEX. A standardized instrument for the diagnosis of mental disorder in the elderly with special reference to the early detection of dementia. Br J Psychiatry 1986; 149: 698-709. 16. Folstein MF, Folstein SE, McHugh PR. "Mini-mental State": a practical method for grading the cognitive state of patients for the clinician. J Psychiat Res 1975; 12: 189-98. 17. Gustafson L, Edvinsson L, Dahlgren N, et al. Intravenous physostigmine treatment of Alzheimer’s disease evaluated by psychometric testing, regional cerebral blood flow (rCBF) measurement, and EEG. Psychopharmacology 1987; 93: 31-35. 18. Battistin L, Pizzolato G, Dam M, et al. Single-photon emission computed tomography studies with 99mTc-hexamethylpropyleneamine oxime in dementia: effects of acute administration of L-acetylcarnitine. Eur Neurol 1989; 29: 261-65. 19. Estrada C, Hamel E, Krause DN. Biochemical evidence for cholinergic innervation of intracerebral blood vessels. Brain Res 1983; 266: 261-70.

Similarities in intestinal humoral immunity in dermatitis herpetiformis without enteropathy and in coeliac disease

Intestinal humoral immunity was examined in eight patients with dermatitis herpetiformis and normal jejunal histology (as determined by

quantitative morphometry) on a glutencontaining diet. Jejunal aspirate was taken at the time of jejunal biopsy, and levels of total immunoglobulins (IgA, IgM, IgG) and specific antibody to gliadin and two other dietary proteins, betalactoglobulin and ovalbumin, were measured. The pattern of secretory immune responses in the dermatitis herpetiformis patients was similar to that in twenty-six patients with untreated coeliac disease—ie, higher than normal concentrations of IgA, IgM, and IgG and high levels of specific antibodies (IgA and IgM) to the three dietary proteins. Serum levels of IgA antigliadin were similar in the dermatitis herpetiformis and control (twenty-eight patients who underwent jejunal biopsy to exclude coeliac disease) groups, and serum levels of IgG antigliadin were intermediate between those of the control and coeliac disease groups. These findings suggest that investigation of gut humoral immunity may provide a diagnostic index of latent coeliac disease. The definition of coeliac disease as a permanent gluten-sensitive enteropathy may have to be revised if the proposed two-stage model is confirmed.

Introduction Coeliac disease probably results from an abnormal intestinal immune response to gluten in genetically predisposed individuals;1 as previously defmed, the gluten-sensitive enteropathy is permanent.:! This definition is now being challenged by the emerging concept of a two-stage model of coeliac disease, based on the postulated existence of latent gluten-sensitive enteropathy or coeliac disease as a fairly common disorder which requires a second insult to allow full evolution of gluten-induced intestinal damage and malabsorption. Latent gluten-sensitive enteropathy was first described in patients with dermatitis herpetiformis, in whom villus flattening developed when they were given additional dietary gluten.3 The latent disorder may also be present in apparently healthy individualsfirst-degree relatives of patients with coeliac disease,s and some patients with diarrhoea who have high jejunal intraepithelial lymphocyte (IEL) counts.6 This disorder could account for the wide variations in the incidence of coeliac disease in place

ADDRESSES. Gastrointestinal Unit, University of Edinburgh, Western General Hospital (S. O’Mahony, MRCPI, Prof A. Ferguson, FRCP) and Department of Dermatology, Royal Infirmary (J. P. Vestey, MRCP), Edinburgh, UK. Correspondence to Dr S. O’Mahony, Gastrointestinal Laboratory, Western General Hospital, Edinburgh EH4

2XU, UK

1488

TABLE I-CLINICAL DETAILS OF PATIENTS WITH DERMATITIS HERPETIFORMIS

C/M ratio = cellobiose/mannitol epithelial cells

ratio.

*Per 100 villus

and

time,’ and the few well-documented cases of acquired gluten-sensitive enteropathy.88 Direct investigation of intestinal humoral immunity in coeliac disease, with jejunal aspirates or whole-gut lavage fluid, have shown an unusual isotype pattern of secretory antibodies.9 Patients with coeliac disease have high levels of total IgM and of IgM antibodies to gliadin and other proteins, as well as high levels of secretory IgA antibody to gliadin and other proteins, though these are less diseasespecific. The abnormality of secretory IgM antibodies persists after many years of treatment with a gluten-free diet; although it is unrelated to enteropathy, it may signal the underlying defect of mucosal immunity in coeliac disease patients. Most patients with dermatitis herpetiformis have a gluten-sensitive enteropathy virtually identical to that of coeliac disease,lo and patterns of serum and secretory antibodies identical to those reported in patients with untreated coeliac disease. In an unpublished study of eight patients with dermatitis herpetiformis and villus flattening, who were on a normal diet (containing gluten), seven had high levels of serum IgA antigliadin and seven high serum IgG antigliadin; in jejunal aspirate six had high IgA antigliadin and all eight had high IgM antigliadin. A minority of patients with dermatitis herpetiformis have completely normal jejunal morphology and function.ll The first descriptions of latent gluten-sensitive enteropathy were of this subgroup within dermatitis herpetiformis patients. We decided to study the intestinal humoral immunity of such patients to determine whether abnormal immunological reactivity, particularly to gliadin, occurs in the absence of enteropathy. Patients and methods We studied eight patients with dermatitis herpetiformis (table i), all of whom had granular IgA at the dermal papillae on direct immunofluorescence of non-lesional skin. All were receiving dapsone to control the rash and all were taking a normal diet. The median daily gluten intake, assessed by a dietitian, was 14 g (range 5-32 g). The jejunal biopsy samples from these patients were normal on subjective histopathological examination, and measurements of villus and crypt lengths,12 IEL counts" on sections stained with haematoxylin and eosin, and activities of brush-border disaccharidase were normal. We studied twenty-eight control patients (fourteen women, fourteen men, median age 35 [range 14-75] years) who had jejunal

biopsy to exclude coeliac disease, no significant pathology was found, and a final diagnosis of functional bowel disease was made. We also studied twenty-six patients with untreated coeliac disease (thirteen women, thirteen men, median age 34 [range 15-74] years)

Levels of antibody to gliadin in

serum

(A)

and

jejunal aspirate

(B). UCD= untreated

coeliac

disease;

herpetiformis, C=control. Horizontal

in whom

DH = normal-biopsy

dermatitis

bars=medians.

jejunal biopsy samples had the features of subtotal or partial villus atrophy. 10-15% of patients with dermatitis herpetiformis have linear, rather than granular, deposits of IgA at the dermal papillae.14 This disorder is not thought to be gluten-sensitive, and jejunal histology is characteristically normal. Leonard et als proposed that the term dermatitis herpetiformis should be reserved for patients with severe

1489

TABLE II-JEJUNAL ASPIRATE LEVELS OF IMMUNOGLOBULINS AND SPECIFIC ANTIBODIES

as percentage of optical density of positive standard †All values in disease groups significantly higher (p < 0-01, Mann-WhItney U) than in controls.

*Expressed

granular IgA deposits. Two patients with linear IgA disease, both with normal jejunal histology, were included in this study. The following investigations were carried out in all the subjects. To assess intestinal permeability we used the cellobiose/mannitol test, as described in detail elsewhere -16the normal urinary cellobiose/mannitol ratio is 0-037 or lower. Jejunal aspirate samples were collected from a point just distal to the duodenojejunal junction, through the tubing of the Crosby capsule, before a biopsy sample was taken. Phenylmethylsulphonylfluoride (Sigma) 100 mmol/1 in 95% alcohol (20 µl per ml aspirate) was added before the samples were divided into several volumes and stored at - 70°C. Levels of antibodies to gliadin, betalactoglobulin, and ovalbumin were measured by enzyme-linked immunosorbent assay (ELISA).9 Serum from a patient with untreated coeliac disease, previously found to have high titres of antibodies of all isotypes to a wide variety of dietary antigens, was used as a reference standard. Plates were read when this sample reached an arbitrary optical density of 1 -0 and readings for test specimens were expressed as percentages of the standard. Total immunoglobulins (IgA, IgM, IgG) were also measured by ELISA. Differences in immunoglobulin and antibody levels between patient groups were analysed by the Mann-Whitney U test.

Results

Despite normal biopsy morphology, the cellobiose/ mannitol ratios were higher than normal in four of the patients with dermatitis herpetiformis, equivocal in one, and normal in three (table I). All control patients had ratios within the normal range, and all coeliac disease patients ratios above the normal range. As reported previously, the levels of serum IgA and IgG antibodies to gliadin were significantly (both p < 000001) higher in patients with coeliac disease than in controls (see accompanying figure); the IgA antibodies were more disease specific in that there was less overlap between patients and controls. There was no significant difference between the dermatitis herpetiformis and control groups in the serum TABLE III—JEJUNAL ASPIRATE IMMUNOGLOBULIN CONCENTRATIONS AND ANTIGLlADIN LEVELS IN PATIENTS WITH LINEAR

"Expressed

as

IgA DISEASE

percentage of optical density of positive standard

IgA antigliadin level,

but

IgG antigliadin

levels

were

intermediate between those of the coeliac disease (p < 0-05) and control (p < 0’05) groups (figure). The levels of intestinal immunoglobulins and antibody responses were similar in the dermatitis herpetiformis and coeliac disease groups (table II), with high levels of total immunoglobulins (IgA, IgM, IgG) and of IgA and IgM antigliadin in the jejunal aspirate (figure, B; p < 000001 for differences between control group and each of the two disease groups). Of these, IgM antigliadin was more specific, with no overlap between the dermatitis herpetiformis and control groups. High levels of secretory antibody to betalactoglobulin (IgA p < 0001, IgM p < 0-002 for difference from control group) and ovalbumin (IgA p < 0’001, IgM p < 0-0002) were also found. There was no significant difference in antibody or immunoglobulin levels between the coeliac disease and dermatitis herpetiformis groups.

Jejunal immunoglobulin and antibody levels in the two patients with linear IgA disease are shown in table III; their pattern of secretory immunity was similar to that of controls. Discussion The characteristic pattern of intestinal humoral immunity in coeliac-disease patients with enteropathy persists after mucosal recovery on a gluten-free diet. We have now shown identical profiles of immunoglobulin concentrations and secretory antibodies in jejunal aspirates of patients with dermatitis herpetiformis and normal jejunal biopsies. These findings strongly suggest that there is a permanent abnormality of mucosal immunity, common to all with coeliac disease or dermatitis herpetiformis, and that when it occurs in a patient with a histologically normal jejunal biopsy, it signals the state of latent coeliac disease. Detailed clinical studies of such patients, with careful measurements of mucosal biopsy specimens taken after periods of gluten loading with withdrawal, will be necessary to prove this

theory. Why

does an intestinal lesion indistinguishable from coeliac disease develop in some patients with dermatitis herpetiformis, whereas others have no evidence of mucosal damage? It has been suggested that the mucosal injury in dermatitis herpetiformis is patchy,17 and that all patients have a degree of enteropathy, but later evidence does not support this suggestion.18 Another theory is that the enteropathy in dermatitis herpetiformis is related to daily gluten intake,19 but we found no such relation in a series of 51 dermatitis herpetiformis patients taking a normal diet.20 Marsh described the spectrum of jejunal biopsy appearances in dermatitis herpetiformis and coeliac disease,8 from the minimum lesion of lymphocyte infiltration through crypt hyperplasia to a flat mucosa. He also showed that a rise in the number of IEL in the villus epithelium is the first detectable jejunal change after gluten challenge in treated coeliac disease patients and that there may be progression to subtotal villus atrophy after a greater gluten challenge. Some first-degree relatives of patients with coeliac disease have similar IEL infiltration of the jejunal mucosa,21 which may reflect the fact that clinical sensitivity to a given amount of gluten varies enormously within a population. Nevertheless, some of the patients and volunteers in whom enteropathy developed after gluten loading had normal baseline counts of IEL. 4,5 Thus, it seems that the spectrum of potential gluten sensitivity extends to apparently normal jejunal histology-

1490

There is good evidence that the skin lesions in patients with normal-biopsy dermatitis herpetiformis respond to a gluten-free diet,2z and our finding of a coeliac-like pattern of secretory immunity supports this clinical observation. Conversely, patients with linear IgA disease do not respond to such treatment; as expected our two patients with this disorder had normal secretory immunity. Our findings in normal-biopsy dermatitis herpetiformis patients support the view that intestinal antigliadin is not directly involved in an enteropathogenic process. Furthermorethe pattern of secretory immunity in these patients shows the virtual independence of the mucosal and systemic humoral immune systems; some patients with low serum levels of IgA antigliadin had high levels of secretory antibody to gliadin. Studies based on serum have little relevance to events in the intestine in these disorders. Much effort has been directed towards the development of an animal model of gluten-sensitive enteropathy. Our studies with niice 13 showed that immunological sensitisation to gliadin does not trigger the development of a T-cellmediated lesion of the intestine when the diet contains gluten. Additional factors, such as those occurring during intestinal anaphylaxis or a graft-versus-host reaction, were necessary. Enhanced antigen presentation, recruitment of specific T cells in the mucosa, upregulation of the expression of class II antigens, and failure of suppression are all candidate mechanisms for the effects observed. The proposed two-stage model of coeliac disease derives from the confluence of several lines of clinical and experimental work. 3-6,8,21,24-26 The model is that inappropriate immunity to gliadin is quite common; it is genetically restricted, and may be expressed not only in gut and skin, but also in the mouth (recurrent aphthae),24 kidneys (IgA nephropathy) ’25 and joints (some types of arthritis).26 Expression of T-cell-mediated immunity to gliadin in the gut occurs across a spectrum of histological and functional abnormalities. The minimum lesion may appear histologically normal with or without a high count of villus IEL; the fully expressed lesion is a flat mucosa with crypt hyperplasia, typical of coeliac disease. Mucosal immunological sensitisation is an invariable feature of coeliac disease but is not the precipitating factor for the expression of the full intestinal lesion; a second factor drives the enteropathy from latent to overt, either by immunological mechanisms or by direct ancillary effects on enterocytes. Possible factors include an episode of hyperpermeability, nutrient deficiency, a rise in dietary gluten, impaired intraluminal digestion of ingested gluten, adjuvant effects of intestinal infection, and a non-HLAassociated gene. Several factors may act together as the trigger mechanism. We have described a simple approach to studies of secretory immunity that might allow the recognition of latent coeliac disease in likely groups of patients. These groups include first-degree relatives of patients with coeliac disease;21patients with idiopathic oral ulceration ;24 and patients with gluten-sensitive diarrhoea without overt enteropathy.6 Such studies should be complemented by determination of HLA class II markers, measurement of intestinal permeability, and quantitative indices of jejunal

morphology. We thank the nursing staff of the Gastrointestinal Investigation Suite, Western General Hospital; and Mr N. Anderson and Mrs J. Johnstone who provided technical assistance. This study was supported by Fisons Pharmaceuticals (to S. O’M.).

REFERENCES 1. Strober W. An immunological theory of gluten-sensitive enteropathy. In: McNicholl B, McCarthy CF, Fottrel PF, eds. Perspectives in coeliac disease. Lancaster: MTP, 1978: 169-82. 2. Meeuwisse G. Diagnostic criteria in coeliac disease. Acta Paediatr Scand 1970; 59: 461-64. 3. Weinstein WM. Latent celiac sprue. Gastroenterology 1974; 66: 489-93. Ferguson A, Blackwell JN, Barnetson RStC. Effects of additional dietary gluten on the small-intestinal mucosa of volunteers and of patients with dermatitis herpetiformis. Scand J Gastroenterol 1987; 22: 543-49.

4.

RE. Gluten-induced mucosal changes in subjects without overt small-bowel disease. Lancet 1981; i: 517-20. 6. Cooper BT, Holmes GKT, Ferguson R, Thompson RA, Allan RN, Cooke WT. Gluten-sensitive diarrhea without evidence of celiac disease. Gastroenterology 1980; 79: 801-06. 5.

Doherty M, Barry

Logan RFA, Rifkind EA, Busuttil A, Gilmour HM, Ferguson A. Prevalence and "incidence" of celiac disease in Edinburgh and the Lothian region of Scotland. Gastroenterology 1986; 90: 334-42. 8. Marsh MN. Studies of intestinal lymphoid tissue. XI-the immunopathology of cell-mediated reactions in gluten sensitivity and other enteropathies. Scanning Microsc 1988; 2: 1663-84. 9. O’Mahony S, Arranz E, Barton JR, Ferguson A. Dissociation between systemic and mucosal humoral immune responses in coeliac disease. Gut (in press). 10. Fry L, Keir P, McMinn RMH, Cowan JD, Hoffbrand AV. Small intestinal structure and function and haematological changes in dermatitis herpetiformis. Lancet 1967; ii: 729-34. 11. Gawkrodger DJ, McDonald C, O’Mahony S, Ferguson A. Small bowel function in dermatitis herpetiformis. Br J Dermatol 1989; 121 (suppl 34): 49. 12. Ferguson A, Sutherland A, McDonald TT, Allan F. Technique for microdissection and measurement in biopsies of human small intestine. J Clin Pathol 1977; 30: 1068-73. 13. Ferguson A, Murray D. Quantitation of intraepithelial lymphocytes in human jejunum. Gut 1971; 12: 988-94. 14. Chorzelski TP, Jablonska S, Beutner EH, Wilson DB. Linear IgA bullous dermatosis. In: Beutner EH, Chorzelski TP, Kumar V, eds. Immunopathology of the skin, 3rd ed. New York: Wiley, 1987: 407-20. 15. Leonard JN, Haffenden GP, Ring GP, Fry L. Linear IgA bullous dermatosis in adults—the St Mary’s view. In: Beutner EH, Chorzelski TP, Kumar V, eds. Immunopathology of the skin, 3rd ed. New York: Wiley, 1987: 421-29. 16. Strobel S, Brydon WG, Ferguson A. Cellobiose/mannitol sugar permeability test complements biopsy histopathology in clinical investigation of the jejunum. Gut 1984; 25: 1241-46. 17. Brow JR, Parker F, Weinstein WM, Rubin CE. The small intestinal mucosa in dermatitis herpetiformis 1: severity and distribution of the small intestinal lesion and associated malabsorption. Gastroenterology 7.

1971; 66: 355-61.

Gawkrodger DJ, Blackwell JN, Gilmour HM, Rifkind EA, Heading RC, Barnetson RStC. Dermatitis herpetiformis: diagnosis, diet and demography. Gut 1984; 25: 151-57. 19. Andersson H, Bjorkman AC, Gillberg R, Kastrup W, Mobacken H, Stockbrugger R. Influence of the amount of dietary gluten on gastrointestinal morphology and function in dermatitis herpetiformis. 18.

20.

Hum Nutr Clin Nutr 1984; 38: 279-85. Gawkrodger DJ, McDonald C, O’Mahony S, Ferguson A. Small intestinal function and dietary status in dermatitis herpetiformis. Gut

(in press). 21. Marsh MN, Bjarnason I, Shaw J, Ellis A, Baker R, Peters TJ. Studies of intestinal lymphoid tissue. XIV—HLA status, mucosal morphology, permeability and epithelial lymphocytes in first degree relatives of patients with coeliac disease. Gut 1990; 31: 32-36. 22. Reunala T, Blomqvist K, Tarpila S, Halme H, Kangas K. Gluten free diet in dermatitis herpetiformis. 1. Clinical response of skin lesion in 81 patients. Br J Dermatol 1977; 97: 473-80. 23. Troncone R, Ferguson A. Induction of intestinal cell-mediated immunity to gliadin in mice. In: Proceedings of the International Coeliac Symposium, St Bartholomew’s Hospital, London, 1988: 14. 24. Wray D. Gluten-sensitive recurrrent aphthous stomatitis. Dig Dis Sci 1981; 26: 737-40. 25. Laurent J, Branellec A, Rostoker G, et al. An increase in circulating IgA antibodies to gliadin in IgA mesangial glomerulonephritis. Am J Nephrol 1987; 7: 178-83. 26. O’Farrelly C, Marten D, Melcher D, et al. Association between villous atrophy in rheumatoid arthritis and a rheumatoid factor and gliadinspecific IgG. Lancet 1988; ii: 819-22.