Dermatitis Herpetiformis: Pathognomonic Transglutaminase IgA Deposits in the Skin and Excellent Prognosis on a Gluten-free Diet

Ac a De m Vene eol 95 REVIEW ARTICLE Ac a De m Vene eol 2015; 95: 917–922 © 2015 The Au ho s. doi: 10.2340/00015555-2162 Jou nal Compila ion © 2015 Ac a De ma o-Vene eologica. ISSN 0001-5555 De ma i is he pe i o mis (DH) is an i chy, blis e ing skin disease wi h si es o p edilec ion a he elbows, knees and bu ocks. Al hough DH is mos ly asymp oma ic, all pa ien s exhibi small bowel illous a ophy o a leas coeliac- ype in lamma o y changes. Deposi ion o im- munoglobulin A (IgA) in he papilla y de mis is a key diagnos ic ea u e o DH. Epide mal ansglu aminase (TG3) is he an igen o IgA deposi ed in he skin, and issue ansglu aminase (TG2) is he an igen o IgA de- posi ed in he small bowel mucosa. Clinically silen , bu immunologically ac i e coeliac disease in he gu appea s o esul in IgA TG3 an ibody complexes agg ega ed in o DH skin. The p e alence o DH in no he n Eu ope is high (30–75/100,000), bu i s incidence is dec easing, possibly due o inc eased ecogni ion o subclinical co- eliac disease. The ash and small bowel heal on a glu en- ee die , which is a li e-long ea men . The isk o non- Hodgkin’s lymphoma is inc eased, bu in pa ien s wi h DH who adhe e s ic ly o a glu en- ee die long- e m p ognosis is excellen . Key wo ds: de ma i is he pe i o - mis; coeliac disease; glu en- ee die ; ansglu aminase au oan ibodies; immunoglobulin A deposi s. Accep ed Jun 3, 2015; Epub ahead o p in Jun 10, 2015 Ac a De m Vene eol 2015; 95: 917–922. Timo Reunala, Medical School, Uni e si y o Tampe e, FIN-33014 Tampe e, Finland. E-mail: [email p o ec ed] De ma i is he pe i o mis (DH) was i s desc ibed as a clinical en i y by Louis Duh ing in 1884, and was associa- ed wi h coeliac disease (CD) in 1966 when en e opa hy was disco e ed in he small bowel o DH pa ien s (1, 2). Subsequen ly, he blis e ing ash wi h si es o p edilec ion on he elbows, knees and bu ocks was ound o espond o a glu en- ee die (GFD) (3, 4). Fu he e idence o he ela ionship be ween DH and CD came om immuno- gene ic and amily s udies. Bo h diseases we e ound o ha e he same s ong associa ion wi h human leukocy e an igen (HLA) DQ2, which is ca ied by up o 100% o pa ien s wi h DH (5, 6). Mo eo e , DH and CD clus e in he same amilies, and monozygo ic wins (1 wi h DH and he o he wi h CD) ha e been epo ed (7, 8). Demons a ion o g anula immunoglobulin A (IgA) deposi s in he unin ol ed skin is an easy way o di- agnose DH, whe eas small bowel biopsy is needed o con i m he diagnosis o CD (9–11). In bo h diseases pa ien s ha e IgA class ci cula ing an ibodies, i s de- sc ibed o e iculin, hen o endomysium (EmA) and, inally, o issue ansglu aminase (TG2; 12, 13). The b eak h ough in CD esea ch was he inding by Schup- pan and co-wo ke s in 1997 ha issue ansglu aminase (TG2) enzyme, which is p esen in he gu mucosa, was he au o-an igen (14). In 2002 Sá dy e al. (15) p esen ed e idence ha epide mal ansglu aminase (TG3) was an au o-an igen o IgA deposi ed in DH skin. The ea e , IgA TG2 deposi s ha e been shown o occu in he small bowel mucosa o mos un ea ed pa ien s ei he wi h CD o DH (16, 17). Se e al expe e iews (18–22), 2 na ional guidelines (23, 24), a con inuous medical examina ion (25), and a clinical p ac ice a icle (26) ha e been published on DH in ecen yea s. The p esen e iew ocuses on he la es esea ch indings and p o ides u he e idence ha DH is a speci ic skin mani es a ion o CD. IMMUNOPATHOGENESIS OF DERMATITIS HERPETIFORMIS The IgA deposi s in he papilla y de mis o DH skin ha e long been suspec ed o de i e om he gu . Following disco e y o he glu en-sensi i e en e opa hy, Seah e al. (27) p oposed, in 1971, ha IgA o igina es om he gu in glu en-an iglu en immune complexes. These a e hen ap- ped in he skin as a esul o c oss- eac i i y wi h deposi ed e iculin an ibodies. Al hough glu en-an iglu en immune complexes could be ound in he se um o pa ien s wi h DH, s udies did no e eal any glu en o glu en an ibodies in he skin (28). In 2002, Sá dy e al. (15) demons a ed ha he an igen o deposi ed IgA was TG3 enzyme. They used dual immunos aining o he p esence o IgA, and showed ha TG3 and IgA co-localized. This impo an inding was con i med by Donaldsson e al. (29). Sá dy e al. (15) also showed ha , al hough pa ien s wi h CD also p esen ed wi h IgA class TG3 an ibodies, he an i- bodies in pa ien s wi h DH ecognized TG3 selec i ely and wi h high a idi y. TG3 and TG2 a e closely ela ed and show a high deg ee o sequence conse a ion, mos ly wi hin enzyma ically ele an domains (20). The e o e, c oss- eac i i y o he an ibodies agains TG3 and TG2 is no su p ising. The abili y o TG2 o deamida e and c oss-link glu en pep ides is essen ial o he p oduc ion De ma i is He pe i o mis: Pa hognomonic T ansglu aminase IgA Deposi s in he Skin and Excellen P ognosis on a Glu en- ee Die Timo REUNALA1,2, Teea T. SALMI1,2 and Kaisa HERVONEN1,2 1Depa men o De ma ology, Tampe e Uni e si y Hospi al, and 2School o Medicine, Uni e si y o Tampe e, Tampe e, Finland 918 T. Reunala e al. o TG2 au oan ibodies in CD. TG3 o ms glu en pep ide complexes less e icien ly, which could be a eason o he di e en au o-an ibody esponse in DH (30). Whe he he high a ini y IgA an ibodies o TG3 in pa ien s wi h DH a ise agains TG3 as a p ima y an igen o a e he esul o epi ope sp eading is s ill an open ques ion, because TG3 p o ein has no been de ec ed in he small bowel simila ly o he TG2 enzyme (15, 20). An immunopa hogenesis o DH, s a ing om subclinical CD in he gu and e ol ing o immune complex deposi ion o high a idi y IgA TG3 an ibodies oge he wi h TG3 enzyme in he papilla y de mis, is shown in Fig. 1. One impo an poin is ha , in no mal skin, TG3 is ex- p essed in he ke a inocy e laye s and no in he papilla y de mis whe e he IgA deposi s a e loca ed in DH (10, 15, 29). Hence, he IgA p ecipi a es could be immune com- plexes con aining TG3, which accumula e speci ically in he papilla y de mis (20). Suppo ing his, TG3 deposi s ha e also been obse ed in cu aneous essels in DH skin (31). Recen ly, Taylo e al. (32) showed ha TG3 p esen in IgA agg ega es in DH skin is enzyma ically ac i e and can bind soluble ib inogen. This i s well wi h an ea lie s udy showing ib inogen a he same si e as he IgA de- posi s in he unin ol ed skin (33). Mo eo e , e ol ing DH blis e s show ma ked ib in deposi ion and up egula- ion o u okinase ype plasminogen ac i a o , sugges ing enhanced ib inolysis (34, 35). The blis e ing ash in DH has si es o p edilec ion a he knees, elbows and bu ocks, al hough IgA agg ega es a e also deposi ed in si es ha a e ne e in ol ed in lesion o ma ion (10, 29). The mos likely explana ion o his unique dis ibu ion o he ash in ol es he in luence o local ac o s, such as p essu e and s e ching (Fig. 1). I is possible ha di ec ac i a- ion o TG3 in de mal agg ega es by mechanical o ce, simila ly o TG2 ac i a ion in ascula walls, could be an ini ia o leading o blis e o ma ion (36). This would esul in elease o ib inogen om he agg ega es (32, 33), which, besides being a clo ing ac o , is an in lam- ma o y p o ein capable o a ac ing T cells, neu ophils and mac ophages, all o which ha e been shown o in lux in o he de eloping DH lesions (34, 37). One in e es ing ques ion is why IgA deposi s pe sis in DH skin in pa ien s on a GFD long a e he ash has become asymp oma ic and ci cula ing TG3 au o- an ibodies ha e disappea ed om he se um (38–40). In a ecen s udy (41) we ound 3 pa ien s wi h DH wi h IgA deposi s, despi e he ac ha hey had been asymp oma ic on a s ic GFD o a mean o 8 yea s. The eason o he unexpec edly long pe sis ence o IgA TG3 agg ega es seems o be he igh binding o he ex acellula ma ix o he papilla y de mis. Ac i e c oss-linking is u he subs an ia ed by he obse a ion ha TG3 e ains a leas pa o i s enzyma ic ac i i y in DH skin (32). In con as o he long ime aken o IgA deposi s o disappea a e wi hd awal o glu en, hese seem o eappea mo e apidly, wi hin one yea on glu en challenge (42). Fu he glu en challenge s udies, ocusing on he eappea ance o IgA TG3 agg ega es in he skin and TG2 agg ega es in he small bowel, would p o ide impo an knowledge abou he ini ial e en s in he immunopa hogenesis o DH. CLINICAL PRESENTATION AND DIAGNOSIS OF DERMATITIS HERPETIFORMIS DH can appea a any age. The age o he younges pa ien in ou Tampe e se ies is 3 yea s and he oldes 84 yea s (43). Mean age a diagnosis o DH was 43 yea s, bo h in Fig. 1. Immunopa hogenesis o de ma i is he pe i o mis s a ing om subclinical coeliac disease in he gu . Small bowel mucosa: gliadin pep ides a e modi ied by issue ansglu aminase (TG2) enzyme and ecognized by human leukocy e an igen (HLA) DQ2/8- posi i e an igen p esen ing cells a e which B cells/plasma cells p oduce immunoglobulin A (IgA) an ibodies o gliadin pep ides and TG2. Blood: ci cula ing IgA class an ibodies o gliadin, TG2 and epide mal ansglu aminase (TG3) a e p esen . High-a idi y TG3 an ibodies may a ise om TG2 an ibodies by epi ope sp eading. Unin ol ed skin: immune complexes consis ing o IgA TG3 an ibodies and TG3 enzyme a e deposi ed in he papilla y de mis. S e ching/p essu e, o possibly o he ac o s, ac i a e deposi ed TG3 enzyme, which is capable o binding ib inogen. This and/o o he ac o s a ac in lamma o y cells, such as T lymphocy es and polymo phonuclea leucocy es (PMNs), o in lux in o he papilla y de mis. These cells sec e e in lamma o y media o s, such as a ious cy okines and p o eases, which inally cause subepide mal blis e o ma ion. Ac a De m Vene eol 95 919 De ma i is he pe i o mis: ansglu aminase IgA deposi s and glu en- ee die men and women, bu i has inc eased signi ican ly du ing ecen yea s. Simila mean ages ha e been epo ed in DH se ies om Eu ope, No h Ame ica and Asia. Recen DH se ies om I aly (44) include many child en, and he mean age o he pa ien s is only 32 yea s. I is possible ha DH appea s in childhood mo e o en in sou he n han no he n Eu ope, since in Finland childhood DH is a e, comp ising only 4% o all pa ien s (45). Males sligh ly o ma kedly p edomina e in adul s wi h DH (Table I), whe eas he op- posi e is ue in child en (45, 46) and in CD (11, 47). The ypical si es o p edilec ion o DH a e he elbows, knees and bu ocks (Fig. 2). In addi ion, he uppe back, abdomen, g oin, axillae, scalp and ace can be a ec ed, bu o al lesions a e a e (48). The ash is polymo phic wi h small blis e s (Fig. 2). These a e, howe e , o en e oded and c us ed because o in ense i ch and sc a ching. Palma o ac al pu pu ic lesions may appea on he hands (49). The p esen a ion and ac i i y o he ash a ies g ea ly om pa ien o pa ien , bu comple e emission is in e- quen on a no mal, glu en-con aining die . The clinical pic u e is o en highly sugges i e o DH al hough, linea IgA disease is always a diag- nos ic p oblem (50). The symp oms and signs o mild DH a e easily masked i he pa ien has a concomi an i chy skin diso de , such as a opic de ma i is (45). The bu ning i ch el du ing he de elopmen o blis e s is, howe e , usually se e e enough o aise suspicion o DH. The ideal me hod o diagnosis o DH is a di ec immuno luo escence biopsy o una ec ed skin in close p oximi y o an ac i e lesion. This e eals pa hogno- monic g anula IgA deposi s a he de mo-epide mal junc ion (Fig. 2), and he diagnosis o DH should no be made wi hou his inding (10, 18, 19). SMALL BOWEL AND SEROLOGICAL FINDINGS IN DERMATITIS HERPETIFORMIS Pa ien s wi h DH a ely p esen wi h ab up gas oin- es inal symp oms o signs o malabso p ion (18, 19, 51). Consis en wi h CD, app oxima ely 70% o pa ien s wi h DH ha e illous a ophy and c yp hype plasia in small bowel mucosal samples om uppe gas oin es inal en- doscopy (18, 19, 51; Table I). The emaining pa ien s ha e ma ke s o ea ly-s age CD, such as inc eased densi y o gamma/del a posi i e in aepi helial T lymphocy es (52, 53). Fu he mo e, in 1988, Ká pá i e al. (54) showed ha child en wi h DH had IgA deposi s in he p oximal jeju- num. In 2004, Ko ponay-Szabó e al. (16) epo ed ha he in es inal IgA deposi s we e di ec ed agains TG2, in bo h CD and DH. Subsequen s udies showed ha almos all un ea ed pa ien s wi h CD and 80% o pa ien s wi h DH ha e TG2- a ge ed IgA deposi s in he small bowel mucosa, and ha hese deposi s a e glu en-dependen (17, 55). Pa ien s wi h un ea ed CD o DH p esen wi h ci cula ing IgA au oan ibodies. An i- e iculin an ibody es was he i s CD-speci ic es and he EmA es he second (12, 56). When TG2 was iden i ied as he au o-an igen in CD, an enzyme- linked immunoso ben assay (ELISA) me hod was es ablished o de ec ing IgA-class TG2 an ibodies in he se um (13, 14). TG2 an ibody ELISA es has p o en highly sensi i e and speci ic o CD. In DH TG2 an ibodies a e ound a a somewha Table I. Compa ison o de ma i is he pe i o mis (DH) and coeliac disease (CD) DH CD Sex Sligh ly mo e males Females p edomina e Age a onse Mainly adul s Child en and adul s 1s -deg ee ela i es wi h CD o DH Yes Yes HLA DQ2 95–100% 95% IgA deposi s in he skin 100% (by de ini ion) 0% Small bowel illous a ophy 70% 100% (by de ini ion) IgA deposi s in bowel mucosaa79% 95–100% IgA TG3 an ibodies in se umb86% 24% IgA TG2 an ibodies in se umb86% 92% IgA EmA an ibodies in se umb89% 95% P ognosis on a glu en- ee die Excellen cInc eased all-cause and lymphoma mo ali yd aSalmi e al. (17), bReunala e al. (40), cHe onen e al. (62), dTio e al. (66). HLA DQ2: human leukocy e an igen DQ2; IgA: immunoglobulin A; TG3: epide mal ansglu aminase; TG2: issue ansglu aminase; EmA: endomysium. Fig. 2. De ma i is he pe i o mis. (a) Polymo phic ash wi h exco ia ed blis e s on he elbows and knees. (b) Typical small blis e s on he elbow. (c) G anula immunoglobulin A (IgA) deposi s a he de mo–epide mal junc ion. Di ec immuno luo escence examina ion o unin ol ed skin. Ac a De m Vene eol 95 920 T. Reunala e al. lowe pe cen age han in CD, and posi i e indings a e con ined mos ly o pa ien s wi h small bowel illous a ophy (57). The majo i y o pa ien s wi h un ea ed DH also ha e ci cula ing an ibodies agains TG3, whe eas he equency is much lowe , up o 24%, in pa ien s wi h CD (40, 58, 59; Table I). In pa ien s wi h DH adhe ing o a GFD IgA an ibodies o TG3 dec ease in pa allel wi h TG2 and EmA an ibodies (40). This indica es ha he measu emen o TG3 an ibodies does no o e any ad an age o e he widely used TG2 an ibody assay o moni o ing GFD ea men . GLUTEN-FREE DIET AND LONG-TERM PROGNOSIS OF DERMATITIS HERPETIFORMIS A GFD is he ea men o choice o pa ien s wi h DH, leading o healing o he ash and small bowel en e opa- hy (3, 4, 38). Howe e , i akes se e al weeks o mon hs o he ash o espond o a s ic GFD and, he e o e, 65% o ou pa ien s also s a dapsone (4,4’-diamino- diphenylsul one) ea men . The ini ial daily dose is 25–50 mg, which causes apid elie o i ching and he ash subsides wi hin 2–3 days. I he e is no e ec , he dose is inc eased o 100 mg daily. Dapsone has a dose- ela ed isk o haema ological side-e ec s, such as me haemoglobinaemia and haemolysis, bu hese a e a e in doses below 100 mg daily. Pa ien s on a GFD we e p e iously ad ised o a oid whea , ye, ba ley and oa s. A p esen hey can consume oa s, which has been shown o be non- oxic (60, 61). Long- e m ollow-up o ou pa ien s wi h DH showed ha 98% adhe ed o he GFD, 72% o hem s ic ly (62). The eason o his excellen compliance appea s o be egula isi s o he specialis ou pa ien clinic un by expe ienced de ma ologis s, o a leas 1–2 yea s o longe , un il dapsone could be s opped. Recen ly, we analysed whe he some o ou pa ien s wi h DH could be non- esponsi e o a s ic GFD (41). We ound 7 (1.7%) pa ien s who s ill used dapsone o ac i e ash, al hough hey had been on a s ic GFD o a mean o 16 yea s. In con as o he ash, he small bowel mucosa had eco e- ed. This indica es ha he condi ion in DH is di e en om e ac o y CD, which is associa ed wi h a ious complica ions and occu ence o in es inal lymphoma (63). The ash in DH eappea s on glu en challenge (42), and he e is gene al ag eemen ha adhe ence o a GFD should be li elong. The e a e, howe e , some s udies sugges ing ha a ew child en and adul s wi h DH may go in o emission and ole a e glu en (64, 65). CD is known o inc eased isk o all-cause mo ali y and non-Hodgkin’s lymphoma mo ali y (66). Recen ly, we analysed he mo ali y a e in ou p ospec i ely collec- ed coho o 476 pa ien s wi h DH (62). The pa ien s we e ollowed up o a o al o 9,079 pe son yea s and almos all o hem adhe ed o a GFD. Unexpec edly, he s anda - dized mo ali y a e (SMR 0.70) was signi ican ly educed compa ed wi h he gene al popula ion. P e iously, Lewis e al. (67) s udied 846 pa ien s wi h DH in No ingham, UK, and ound a sligh ly, bu non-signi ican ly, educed mo ali y a e (haza d a io 0.93). Howe e , one- hi d o hei pa ien s had no da a on adhe ence o a GFD, which could be a eason o he non-signi ican mo ali y a e. In DH he isk o non-Hodgkin’s lymphoma is signi- ican ly inc eased (up o 10 imes), bu ollowing a GFD o mo e han 5 yea s seems o be p o ec i e (68–70). In ag eemen wi h his, we ound in ou ecen DH se ies (62), in which almos all pa ien s adhe ed o a GFD, a signi ican ly inc eased lymphoma mo ali y a e du ing he i s 5 yea s o ollow-up, bu no he ea e . A sig- ni ican ly educed mo ali y a e due o ce eb o ascula diseases (SMR 0.38) was also obse ed, which migh be ela ed o he GFD o o he lowe le el o smoking in pa ien s wi h DH. CONCLUSION DH is ela i ely common, especially among people in no he n Eu ope and in U ah, USA (43, 71, 72). I also occu s in Sou h Ame ica, bu is ex emely a e in Asia, and appea s o be non-exis en among na i e people in A ica due o low consump ion o whea and absence o HLA DQ2 (73–76). Two ecen s udies o DH, he i s om Finland wi h 477 pa ien s (43) and he second om he UK wi h 1,160 pa ien s (71) ound p e alences o 75/100,000 and 30/100,000. In e es ingly, he p e alence o DH in he UK is 8 imes lowe han o CD (71). In Finland bo h diseases appea ed a app oxima ely he same equency in he 1980s, whe eas a p esen he p e- alence o DH is 6 imes lowe han ha o CD (43, 77). Recen se ological sc eening s udies ha e documen ed ha he p e alence o CD is as high as 1%, hus o e e y pa ien iden i ied 7–8 emain undiagnosed (11, 47, 78). This la ge pool o undiagnosed, and mos ly subclinical, CD seems o be he basis om which DH e ol es. In suppo o his heo y, we ha e seen pa ien s who ini ially de eloped CD, hen ollowed o did no ollow a GFD, and inally de eloped DH (79). Mo eo e , pa ien s wi h DH equen ly ha e coeliac- ype den al enamel de ec s (80), which de elop ea ly in childhood as a esul o mal- abso p ion o immune al e a ion caused by undiagnosed CD. The dec easing incidence o DH since he 1990s in Finland and he UK, wi h a simul aneous apid inc ease in CD (43, 71), i s ou hypo hesis ha clinically silen CD is a p e equisi e o he de elopmen o DH (18, 81). The ash, wi h IgA deposi s in he skin, is he main ea- u e ha di e en ia es DH om CD, whe eas small bowel indings a e gene ally simila in bo h condi ions. Tha TG3 enzyme is he au o-an igen o IgA deposi s in DH skin and TG2 o IgA deposi s in coeliac mucosa a e impo an obse a ions (15–17, 29). Whe he IgA TG3 agg ega es in DH skin a e in ol ed in he ini ial e en s o blis e o - ma ion is unknown. G anula IgA deposi s a e, howe e , Ac a De m Vene eol 95 921 De ma i is he pe i o mis: ansglu aminase IgA deposi s and glu en- ee die he key diagnos ic ea u e o DH ha di e en ia es his blis e ing disease om all o he de ma ological diso de s. The ea men o choice o DH is a GFD and, i ca e ully ollowed, he long- e m p ognosis is excellen (62). ACKNOWLEDGEMENTS The au ho s would like o hank nu ses Sanna Sällinen and Tiina Kallio o hei help in unning he specialis ou pa ien clinic o he pa ien s wi h DH. P o Raimo Suhonen p o ided he blis e pho og aph o DH. Tiina Rauha i a, PhD, is acknowledged o he help in p epa ing he diag am abou immunopa hogenesis o DH. The s udy was inancially suppo ed by he Compe i i e Resea ch Funding o Tampe e Uni e si y Hospi al (g an 9N062). REFERENCES 1. Ma ks J, Shus e S, Wa son AJ. Small-bowel changes in de ma i is he pe i o mis. Lance 1966; 2: 1280–1282. 2. F y L, Kei P, McMinn RM, Cowan JD, Ho b and AV. Small- in es inal s uc u e and unc ion and haema ological changes in de ma i is he pe i o mis. Lance 1967; 2: 729–733. 3. F y L, Seah PP, Riches DJ, Ho b and AV. Clea ance o skin lesions in de ma i is he pe i o mis a e glu en wi hd awal. Lance 1973; 1: 288–291. 4. Reunala T, Blomq is K, Ta pila S, Halme H, Kangas K. Glu- en- ee die in de ma i is he pe i o mis. I. Clinical esponse o skin lesions in 81 pa ien s. B J De ma ol 1977; 97: 473–480. 5. Spu kland A, Ing a sson G, Falk ES, Knu sen I, Sollid LM, Tho sby E. De ma i is he pe i o mis and celiac disease a e bo h p ima ily associa ed wi h he HLA-DQ (alpha 1*0501, be a 1*02) o he HLA-DQ (alpha 1*03, be a 1*0302) he e- odime s. Tissue An igens 1997; 49: 29–34. 6. Balas A, Vica io JL, Zamb ano A, Acuña D, Ga cía-No o D. Absolu e linkage o celiac disease and de ma i is he - pe i o mis o HLA-DQ. Tissue An igens 1997; 50: 52–56. 7. Reunala T. Incidence o amilial de ma i is he pe i o mis. B J De ma ol 1996; 134: 394–398. 8. He onen K, Ka ell K, Holopainen P, Collin P, Pa anen J, Reunala T. Conco dance o de ma i is he pe i o mis and celiac disease in monozygous wins. J In es De ma ol 2000; 115: 990–993. 9. an de Mee JB. G anula deposi s o immunoglobulins in he skin o pa ien s wi h de ma i is he pe i o mis. An im- muno luo escen s udy. B J De ma ol 1969; 81: 493–503. 10. Zone JJ, Meye LJ, Pe e sen MJ. Deposi ion o g anula IgA ela i e o clinical lesions in de ma i is he pe i o mis. A ch De ma ol 1996; 132: 912–918. 11. Guj al N, F eeman HJ, Thomson AB. Celiac disease: p e- alence, diagnosis, pa hogenesis and ea men . Wo ld J Gas oen e ol 2012; 18: 6036–6059. 12. Hälls öm O. Compa ison o IgA-class e iculin and endo- mysium an ibodies in coeliac disease and de ma i is he pe- i o mis. Gu 1989; 30: 1225–1232. 13. Die e ich W, Laag E, Schöppe H, Vol a U, Fe guson A, Gille H, e al. Au oan ibodies o issue ansglu aminase as p edic o s o celiac disease. Gas oen e ology 1998; 115: 1317–1321. 14. Die e ich W, Ehnis T, Baue M, Donne P, Vol a U, Riecken EO, e al. Iden i ica ion o issue ansglu aminase as he au oan igen o celiac disease. Na Med 1997; 3: 797–801. 15. Sá dy M, Ká pá i S, Me kl B, Paulsson M, Smy h N. Epi- de mal ansglu aminase (TGase 3) is he au oan igen o de ma i is he pe i o mis. J Exp Med 2002; 195: 747–757. 16. Ko ponay-Szabó IR, Hal unen T, Szalai Z, Lau ila K, Ki- ály R, Ko ács JB, e al. In i o a ge ing o in es inal and ex ain es inal ansglu aminase 2 by coeliac au oan ibodies. Gu 2004; 53: 641–648. 17. Salmi TT, He onen K, Lau ila K, Collin P, Mäki M, Koski- nen O, e al. Small bowel ansglu aminase 2-speci ic IgA deposi s in de ma i is he pe i o mis. Ac a De m Vene eol 2014; 94: 393–397. 18. Reunala TL. De ma i is he pe i o mis. Clin De ma ol 2001; 19: 728–736. 19. F y L. De ma i is he pe i o mis: p oblems, p og ess and p ospec s. Eu J De ma ol 2002; 12: 523–531. 20. Ká pá i S. De ma i is he pe i o mis: close o un a elling a disease. J De ma ol Sci 2004; 34: 83–90. 21. Zone JJ. Skin mani es a ions o celiac disease. Gas oen e- ology 2005; 128: S87–91. 22. Ká pá i S. De ma i is he pe i o mis. Clin De ma ol 2012; 30: 56–59. 23. Cap oni M, An iga E, Melani L, Fabb i P. Guidelines o he diagnosis and ea men o de ma i is he pe i o mis. J Eu Acad De ma ol Vene eol 2009; 23: 633–638. 24. Ingen-Housz-O o S, Joly P, Be na d P, Bedane C, P os C. De ma i is he pe i o mis. Guidelines o he diagnosis and ea men . Ann De ma ol Vene eol 2011; 138: 271–273. 25. Bolo in D, Pe onic-Rosic V. De ma i is he pe i o mis. Pa II. Diagnosis, managemen , and p ognosis. J Am Acad De - ma ol 2011; 64: 1027–1033. 26. Jakes AD, B adley S, Donle y L. De ma i is he pe i o mis BMJ 2014; 348: g2557. 27. Seah PP, F y L, Ho i and AV, Hollbo ow EJ. Tissue an ibo- dies in de ma i is he pe i o mis and coeliac disease. Lance 1971; 2: 834–836. 28. Ka z SI, S obe W. The pa hogenesis o de ma i is he pe i- o mis. J In es De ma ol 1978; 70: 63–75. 29. Donaldson MR, Zone JJ, Schmid LA, Taylo TB, Neuhausen SL, Hull CM, e al. Epide mal ansglu aminase deposi s in pe ilesional and unin ol ed skin in pa ien s wi h de ma i is he pe i o mis. J In es De ma ol 2007; 127: 1268–1271. 30. S amnaes J, Do um S, Fleckens ein B, Aeschlimann D, Sol- lid LM. Glu en T cell epi ope a ge ing by TG3 and TG6; implica ions o de ma i is he pe i o mis and glu en a axia. Amino Acids 2010; 39: 1183–1191. 31. P eisz K, Sá dy M, Ho a h A, Ká pá i S. Immunoglobulin, complemen and epide mal ansglu aminase deposi ion in he cu aneous essels in de ma i is he pe i o mis. J Eu Acad De ma ol Vene eol 2005; 19: 74–79. 32. Taylo TB, Schmid LA, Meye LJ, Zone JJ. T ansglu amina- se 3 p esen in he IgA agg ega es in de ma i is he pe i o mis skin is enzyma ically ac i e and binds soluble ib inogen. J In es De ma ol 2015; 135: 623–625. 33. Ha enden GP, Blenkinsopp WK, Ring NP, Wojna owska F, F y L. The po assium iodide pa ch es in he de ma i is he pe i o mis in ela ion o ea men wi h a glu en- ee die and dapsone. B J De ma ol 1980; 103: 313–317. 34. Rei amo S, Reunala T, Kon inen YT, Saksela O, Salo OP. In lamma o y cells, IgA, C3, ib in and ib onec in in skin lesions in de ma i is he pe i o mis. B J De ma ol 1981; 105: 167–177. 35. Ai ola K, Vaalamo M, Reunala T, Saa ialho-Ke e UK. En- hanced exp ession o in e s i ial collagenase, s omelysin-1, and u okinase plasminogen ac i a o in lesions o de ma i is he pe i o mis. J In es De ma ol 1995; 105: 184–189. 36. Huelsz-P ince G, Belkin AM, VanBa el E, Bakke EN. Ac- i a ion o ex acellula ansglu aminase 2 by mechanical o ce in he a e ial wall. J Vasc Res 2013; 50: 383–395. 37. Ga ioch JJ, Bake BS, Leona d JN, F y L. T lymphocy es in lesional skin o pa ien s wi h de ma i is he pe i o mis. B J De ma ol 1994; 131: 822–826. 38. Ga ioch JJ, Lewis HM, Sa gen SA, Leona d JN, F y L. 25 yea s’ expe ience o a glu en- ee die in he ea men o de ma i is he pe i o mis. B J De ma ol 1994; 131: 541–545. 39. F y L, Ha enden G, Wojna owska F, Thompson BR, Seah PP. IgA and C3 complemen in he unin ol ed skin in de - ma i is he pe i o mis a e glu en wi hd awal. B J De ma ol 1978; 99: 31–37. Ac a De m Vene eol 95 922 T. Reunala e al. 40. Reunala T, Salmi TT, He onen K, Lau ila K, Kau iainen H, Collin P, e al. IgA an iepide mal ansglu aminase an ibodies in de ma i is he pe i o mis: a signi ican bu no comple e esponse o a glu en- ee die ea men . B J De ma ol 2015; 172: 1139–1141. 41. He onen K, Salmi T, Ilus T, Paasiki i K, Vo nanen M, Lau ila K, e al. De ma i is he pe i o mis e ac o y o glu en ee die a y ea men . Ac a De m Vene eol 2015: Jun 18 [Epub ahead o p in ]. 42. Leona d J, Ha enden G, Tucke W, Unswo h J, Swain F, McMinn R, e al. Glu en challenge in de ma i is he pe i o - mis. N Engl J Med 1983; 308: 816–819. 43. Salmi TT, He onen K, Kau iainen H, Collin P, Reunala T. P e alence and incidence o de ma i is he pe i o mis: a 40- yea p ospec i e s udy om Finland. B J De ma ol 2011; 165: 354–359. 44. An iga E, Ve delli A, Calab ò A, Fabb i P, Cap oni M. Clinical and immunopa hological ea u es o 159 pa ien s wi h de ma i is he pe i o mis: an I alian expe ience. G I al De ma ol Vene eol 2013; 148: 163–169. 45. He onen K, Salmi TT, Ku ppa K, Kaukinen K, Collin P, Reunala T. De ma i is he pe i o mis in child en: a long- e m ollow-up s udy. B J De ma ol 2014; 171: 1242–1243. 46. E maco a E, P ampolini L, T ibbia G, Pezzoli G, Gelme i C, Cucchi G, e al. Long- e m ollow-up o de ma i is he pe- i o mis in child en. J Am Acad De ma ol 1986; 15: 24–30. 47. Mooney PD, Hadji assiliou M, Sande s DS. Coeliac disease. BMJ 2014; 348: g1561. 48. Pa inen P, Hie anen J, Malms öm M, Reunala T, Sa ilah i E. Iodine and gliadin challenge on o al mucosa in de ma i is he pe i o mis. Ac a De m Vene eol 2002; 82: 86–89. 49. Tu H, Pa men ie L, S iege M, Spanou Z, Ho n M, Bel ami- nelli H, e al. Ac al pu pu a as leading clinical mani es a ion o de ma i is he pe i o mis: epo o wo adul cases wi h a e iew o he li e a u e. De ma ology 2013; 227: 1–4. 50. Smi h EP, Zone JJ. De ma i is he pe i o mis and linea IgA bullous de ma osis. De ma ol Clin 1993; 11: 511–526. 51. Gawk odge DJ, Blackwell JN, Gilmou HM, Ridkind EA, Heading RC, Ba ne son RS. De ma i is he pe i o mis: diag- nosis, die and demog aphy. Gu 1984:25: 151–157. 52. Sa ilah i E, Reunala T, Mäki M. Inc ease o lymphocy es bea ing he gamma/del a T cell ecep o in he jejunum o pa- ien s wi h de ma i is he pe i o mis. Gu 1992; 33: 206–211. 53. Jä inen TT, Kaukinen K, Lau ila K, Ky önpalo S, Rasmus- sen M, Mäki M, e al. In aepi helial lymphocy es in celiac disease. Am J Gas oen e ol 2003; 98: 1332–1337. 54. Ká pá i S, Kosnai I, Tö ök E, Ko acs JB. Immunoglobulin A deposi ion in he jejunal mucosa o child en wi h de ma i is he pe i o mis. J In es De ma ol 1988; 91: 336–339. 55. Koskinen O, Collin P, Lind o s K, Lau ila K, Mäki M, Kau- kinen K. Use ulness o small-bowel mucosal ansglu ami- nase-2 speci ic IgA deposi s in he diagnosis and ollow-up o celiac disease. J Clin Gas oen e ol 2010; 44: 483–488. 56. Ladinse B, Rossipal E, Pi schiele K. Endomysium an ibo- dies in coeliac disease: an impo ed me hod. Gu 1994; 35: 776–778. 57. Die e ich W, Laag E, B uckne -Tude man L, Reunala T, Ka - pa i S, Zagoni T, e al. An ibodies o issue ansglu aminase as se ologic ma ke s in pa ien s wi h de ma i is he pe i o - mis. J In es De ma ol 1999; 113: 133–136. 58. Jaskowski TD, Hamblin T, Wilson AR, Hill HR, Book LS, Meye LJ, e al. IgA an i-epide mal ansglu aminase an ibo- dies in de ma i is he pe i o mis and pedia ic celiac disease. J In es De ma ol 2009; 129: 2728–2730. 59. Bo oni G, Biagi F, Ciocca O, Vassallo C, Ca ugno A, Cananzi R, e al. IgA an i-epide mal ansglu aminase au- oan ibodies: a sensible and sensi i e ma ke o diagnosis o de ma i is he pe i o mis in adul pa ien s. J Eu Acad De ma ol Vene eol 2013; 27: 836–841. 60. Ha dman CM, Ga ioch JJ, Leona d JN, Thomas HJ, Walke MM, Lo an JE, e al. Absence o oxici y o oa s in pa ien s wi h de ma i is he pe i o mis. N Engl J Med 1997; 337: 1884–1887. 61. Reunala T, Collin P, Holm K, Pikka ainen P, Mie inen A, Vuol eenaho N, e al. Tole ance o oa s in de ma i is he pe- i o mis. Gu 1998; 43: 490–493. 62. He onen K, Alakoski A, Salmi TT, Helako pi S, Kau iai- nen H, Kaukinen K, e al. Reduced mo ali y in de ma i is he pe i o mis: a popula ion-based s udy o 476 pa ien s. B J De ma ol 2012; 167: 1331–1337. 63. Rubio-Tabia A, Mu ay JA. Classi ica ion and managemen o e ac o y celiac disease. Gu 2010; 59: 547–557. 64. Ba della MT, F edella C, T o a o C, E maco a E, Ca alli R, Saladino V, e al. Long- e m emission in pa ien s wi h de ma i is he pe i o mis on a no mal die . B J De ma ol 2003; 149: 968–971. 65. Paek SY, S einbe g SM, Ka z SI. Remission in de ma i is he - pe i o mis: a coho s udy. A ch De ma ol 2011; 147: 301–305. 66. Tio M, Cox MR, Eslick GD. Me a-analysis: coeliac disease and he isk o all-cause mo ali y, any malignancy and lym phoid malignancy. Alimen Pha macol The 2012; 35: 540–551. 67. Lewis NR, Logan RF, Hubba d RB, Wes J. No inc ease in isk o ac u e, malignancy o mo ali y in de ma i is he - pe i o mis: a coho s udy. Alimen Pha macol The 2008; 27: 1140–1147. 68. Lewis HM, Reunala TL, Ga ioch JJ, Leona d JN, F y JS, Collin P, e al. P o ec i e e ec o glu en- ee die agains de elopmen o lymphoma in de ma i is he pe i o mis. B J De ma ol 1996; 135: 363–367. 69. He onen K, Vo nanen M, Kau iainen H, Collin P, Reunala T. Lymphoma in pa ien s wi h de ma i is he pe i o mis and hei i s -deg ee ela i es. B J De ma ol 2005; 152: 82–86. 70. G ainge MJ, Wes J, Solaymani-Doda an M, Ca d TR, Logan RF. The long- e m isk o malignancy ollowing a diagnosis o coeliac disease o de ma i is he pe i o mis: a coho s udy. Alimen Pha macol The 2012; 35: 730–739. 71. Wes J, Fleming KM, Ta a LJ, Ca d TR, C ooks CJ. Incidence and p e alence o celiac disease and de ma i is he pe i o mis in he UK o e wo decades: popula ion-based s udy. Am J Gas oen e ol 2014; 109: 757–768. 72. Smi h JB, Tulloch JE, Meye LJ, Zone JJ. The incidence and p e alence o de ma i is he pe i o mis in U ah. A ch De ma ol 1992; 128: 1608–1610. 73. Cla indo MV, Possebon AT, Soligo EM, Uyeda H, Rua o RT, Empino i JC. De ma i is he pe i o mis: pa hophysiology, clinical p esen a ion, diagnosis and ea men . An B as De ma ol 2014; 89: 865–875. 74. Oha a C, Ishii N, Hamada T, Shimomu a Y, Niizeki H, Daini- chi T, e al. Dis inc cha ac e is ics in Japanese de ma i is he pe i o mis: a e iew o all 91 Japanese pa ien s o e he las 35 yea s. Clin De Immunol 2012: 562168. 75. Zhang F, Yang B, Lin Y, Chen S, Zhou G, Wang G, e al. De ma i is he pe i o mis in China: a epo o 22 cases. J Eu Acad De ma ol Vene eol 2012; 26: 903–907. 76. Hall RP, Cla k RE, Wa d FE. De ma i is he pe i o mis in wo Ame ican blacks: HLA ype and clinical cha ac e is ics. J Am Acad De ma ol 1990; 22: 436–439. 77. Collin P, Huh ala H, Vi a L, Kekkonen L, Reunala T. Di- agnosis o celiac disease in clinical p ac ice: physician’s ale ness o he condi ion essen ial. J Clin Gas oen e ol 2007; 41: 152–156. 78. Fasano A, Ca assi C. Clinical p ac ice. Celiac disease. N Engl J Med 2012; 367: 2419–2426. 79. Salmi TT, He onen K, Ku ppa K, Collin P, Kaukinen K, Reunala T. Coeliac disease e ol ing in o de ma i is he pe- i o mis in pa ien s adhe ing o no mal o glu en- ee die . Scand J Gas oen e ol 2015; 50: 387–392. 80. Aine L, Mäki M, Reunala T. Coeliac- ype den al enamel de ec s in pa ien s wi h de ma i is he pe i o mis. Ac a De m Vene eol 1992; 72: 25–27. 81. Reunala T. De ma i is he pe i o mis: coeliac disease o he skin. Ann Med 1998; 30: 416–418. Ac a De m Vene eol 95