Background: Only limited epidemiologic data are available on autoimmune bullous diseases. Improved diagnostic tools should have led to an increased incidence. To test this hypothesis, all patients with autoimmune bullous disorders who were treated in the Department of Dermatology at the University of Würzburg, Germany, between January 2001 and June 2002 were analysed prospectively.
Patients and Methods: Epidemiologic data of patients diagnosed with an autoimmune bullous disease during this time period were registered and statistically evaluated. Diagnosis was based on the clinical picture and specific immunopathological findings. Only patients from Lower Franconia, a well-defined administrative region of Southern Germany, were included into this study.
Results: During the study period, 41 patients with an autoimmune bullous disease were diagnosed, including 27 with bullous pemphigoid, 4 with pemphigoid gestationis and mucous membrane pemphigoid, 2 with dermatitis herpetiformis and linear IgA disease, and 1 with epidermolysis bullosa acquisita and pemphigus vulgaris, respectively. The highest incidence was calculated for bullous pemphigoid (13.4 per 1 million inhabitants per year) followed by pemphigoid gestationis (2.0) and mucous membrane pemphigoid (2.0). Patients with mucous membrane pemphigoid were found to have the highest mean age at disease onset (76 years) followed by patients with bullous pemphigoid (74 years).
Conclusions: This is the first prospective study on the incidence of autoimmune bullous disorders. Subepidermal blistering autoimmune diseases were shown to be more frequent than previously reported for Central Europe. This is most likely due to improved diagnostic tools for and increased awareness of these diseases.
Autoimmune bullous disorders are characterized by autoantibodies against desmosomal and hemidesmosomal structural proteins of the skin. In the pemphigus group, autoantibodies are directed against desmosomal proteins resulting in an intraepidermal loss of cell contact. This group includes pemphigus vulgaris, pemphigus foliaceus, IgA pemphigus, and paraneoplastic pemphigus. Subepidermal autoimmune bullous disorders are characterized by antibodies against structural proteins of the dermoepidermal junctional. This group includes bullous pemphigoid (BP), mucous membrane pemphigoid, pemphigoid gestationis, linear IgA disease, epidermolysis bullosa acquisita, lichen planus pemphigoides, and anti-p200 pemphigoid. Furthermore dermatitis herpetiformis with antibodies against epidermal/tissue transglutaminase and bullous systemic lupus erythematosus feature subepidermal blistering .
In Germany and France, in 1995, BP was identified as the most common autoimmune bullous disorder with regional incidences of 6.6 and 7.4 newly diagnosed cases/million/year, respectively [2, 3]. The incidence of BP increases with increasing age of patients and to about 150 in those over 90 years . In the present study, we examined if using improved diagnostic tools (ELISA and immunoblotting with cellular and recombinant fragments of the target antigens) leads to a higher incidence of autoimmune bullous disorders. To this end, patients with autoimmune bullous disorders in Lower Franconia who were newly diagnosed at the Department of Dermatology, University of Würzburg, in the time period from January 2001 to June 2002 were analyzed prospectively. In comparison to a retrospective analysis performed 10 years previously in the same region the incidence was double for BP, mucous membrane pemphigoid and epidermolysis bullosa acquisita as well as four-fold for pemphigoid gestationis and linear IgA disease .
Patients and methods
Diagnosis was made on the basis of clinical findings as well as various immunopathological examinations and is depicted in detail in Table 1. The following detection methods for tissue-bound and circulating autoantibodies were employed: direct immunofluorescence (IF) microscopy of a perilesional mucous membrane or skin biopsy, indirect IF microscopy on monkey esophagus as well as indirect IF microscopy and complement-binding test on 1M NaCl split human skin . ELISA with the recombinant 16th non-collagenous domain of collagen type XVII (BP180 NC16A) , ELISA with recombinant desmoglein 1 and desmoglein 3 (MBL, Nagoya, Japan), ELISA with tissue transglutaminase (Phadia, Freiburg, Germany), immunoblot with a concentrated medium of cultured human keratinocytes to detect the soluble ectodomain of collagen type XVII (LAD-1) , immunoblot with the recombinant C-terminal portion of collagen type XVII (BP180 4575; amino acids 1365–1413) as well as immunoblot with an extract of human dermis .
|Diagnosis||Diagnostic criteria||Direct IF1||Indirect IF2||ELISA/immunoblot|
|Bullous pemphigoid n = 27||Blisters/erosions not predominantly on mucous membranes 25/273||IgG > IgA and/or C3 along the BMZ 24/244||Split skin: IgG > IgA on the blister roof 27/27||IgG reactivity to BP180 9/14|
|Pemphigoid gestationis n = 4||As in bullous pemphigoid but association with pregnancy|
|Mucous membrane pemphigoid5 n = 4||Blisters/erosions predominantly on mucous membranes 4/4||IgG and/or IgA and/or C3 along the BMZ 2/4||Split skin:: IgG and/or IgA on the blister roof and/or blister floor 1/4||IgG- and/or IgA reactivity to BP180 or laminin 332 4/4|
|Linear IgA disease n = 2||Blisters/erosions not predominantly on mucous membranes 2/2||IgA > IgG along the BMZ 1/26||Split skin: IgA > IgG on the blister roof 2/2||IgA reactivity to BP180 fragments 1/2|
|Dermatitis herpetiformis n = 2||Urticarial plaques or vesicles 2/2||Granular IgA in the dermal papillae 2/2||Monkey esophagus: IgA to endomysium 2/2||IgA reactivity to tissue transglutaminase 2/2|
|Epidermolysis bullosa acquisita n = 1||Blisters/erosions not predominantly on mucous membranes 1/1||IgG and/or C3 along the BMZ 1/1||Split skin: IgG on the blister floor 1/1||IgG reactivity to collagen type VII 1/1|
|Anti-p200-pemphigoid||Blisters/erosions not predominantly on mucous membranes||IgG and/or C3 along the BMZ||Split skin: IgG on the blister floor||IgG reactivity to a 200 kDa heavy protein in dermal extract|
|Lichen planus pemphigoides||Blisters/erosions and lichenoid papules||IgG and/or C3 along the BMZ||Split skin: IgG on the blister roof||IgG reactivity to 16th non-collagenous domain of BP180|
|Bullous lupus erythematosus||Blisters/erosions; fulfillment of ACR diagnostic criteria for SLE||Band-like deposition of IgG/IgM/IgA/C3 in the BMZ||Split skin: IgG on the blister floor||IgG reactivity to collagen type VII|
|Pemphigus vulgaris n = 1||Blisters/erosions on mucous membranes and/or skin 1/1||IgG and/or C3 intercellularly in epidermis/mucous membrane epithelium 1/1||Monkey esophagus: IgG intercellularly in epithelium 1/1||IgG reactivity to desmoglein 3 1/1|
|Pemphigus foliaceus||Blisters/erosions on skin||as for pemphigus vulgaris||IgG reactivity to desmoglein 1|
|IgA pemphigus||Blisters/erosions on mucous membranes and/or skin||IgA intercellularly in epidermis/mucous membrane epithelium||Monkey esophagus: IgA intercellularly in epithelium||–|
|Paraneoplastic pemphigus7||Blisters/erosions on mucous membranes; neoplasia||IgG and/or C3 intercellularly in epidermis/mucous membrane epithelium||Monkey esophagus: IgG intercellularly in epithelium Monkey bladder: IgG in urothelium||IgG reactivity to desmoglein 1 and/or 3|
The epidemiologic data of patients living in the government district of Lower Franconia were assessed using a standardized questionnaire in the time period from January 2001 to June 2002. The population structure of Lower Franconia for the years 2001/2002 was provided by the Bavarian State Office for Statistics and Data Processing (http:///www.statistik.bayern.de). The Department of Dermatology, University of Würzburg, is the only dermatologic clinic in Lower Franconia.
Forty patients with subepidermal autoimmune bullous disorders as well as one patient with pemphigus vulgaris were diagnosed. The incidence of the individual diseases and the age distribution of the patients are depicted in Table 2. By far the most common disease was BP with a total of 27 patients and an incidence of 13.4 newly diagnosed cases per million population and year. Table 3 depicts the age- and gender-specific incidence of BP.
|Diagnosis1||Number of patients||Mean age at disease manifestation||Incidence2|
|Mucous membrane pemphigoid||4||76.3||2|
|Linear IgA disease||2||31||1|
|Epidermolysis bullosa acquisita||1||36||0.5|
|Age clamss [years]||Number of cases/mean population from 1/01–6/02||Incidence1||Number of cases/mean population from 1/01–6/02||Incidence1||Incidence1|
Incidences of autoimmune bullous disorders published to date have been assessed retrospectively. The present study was performed prospectively. This type of analysis usually provides a more exact assessment of epidemiologic data than retrospective studies. Since no registry exists in German to which all patients with autoimmune bullous disorders are reported, it is probable that the incidences determined in the present study are too low; it can be assumed that several patients from Lower Franconia were also treated in hospitals outside of this region. The relatively short observation period (18 months) is also a possible source of error, in that with exception of BP, a single patient already causes dramatic changes in incidence.
BP is the most common autoimmune bullous disorder in Central Europe and Singapore. In France for the years 1984 to 1992, a regional incidence of 7.4 new cases/million/year was calculated , in Singapore 7.6 . In the present study, we found a total incidence of 13.4 new cases per million population per year. This number is distinctly higher than in France and Singapore (for the years 1988 and 1999) as well as in comparison to a retrospective analysis of patients in Lower Franconia between 1989 and 1994, where the incidence was 6.6 . This increase can most likely be explained by improved diagnostic tools at the Department of Dermatology, University of Würzburg. Further BP patients in prodromal stages and those lacking the classical phenotype were diagnosed. As the incidence of BP clearly rises with increasing age being 189 new cases/million/year in those over 80 years of age, the observed increase in incidence in comparison to the study from 1989 to 1994 can, to a small part, be explained by the increased proportion of the elderly in the population. From 1992 to 2002 the number of over 70-year-olds, who constituted three-fourth of our BP patients, rose by 15% in Lower Franconia. The higher incidence in the 51 to 60-year-old BP patients in comparison to those one decade older (Table 3) is very probably due to the small case number and is not found in larger collectives [4, 10].
The incidence of all subepidermal blistering disorders which was 19.9 in our study, of which BP constituted about two-thirds of the cases, was recently analyzed in Scotland and Great Britain. While the incidence of subepidermal blistering disorders was found to be 14 in the Grampian region in northeast Scotland for the years 1991 to 2002, in Great Britain for 2001 to 2005, it was calculated at 43/million/year [10, 11]. While data collection including establishment of the diagnosis was performed by dermatologists in Lower Franconia, France and Singapore, the data in Great Britain was taken as random samples from a reporting system for general practitioners ; in Scotland they originate from the data base of the Institute for Pathology of the Aberdeen Royal Infirmary . A possibly too high estimate of incidence may have been caused by imprecise diagnostic criteria and double reports. Besides the lack of validated diagnoses by dermatologists doubts of the distinct increase in incidence of BP in Great Britain reported by Langan et al. arise, as the number of positive direct immunofluorescence microscopy findings of the St. John's Institute of Dermatology at St. Thomas Hospital, London, did not rise correspondingly during the same time period [10, 12]. A large advantage of this survey, however, is the inclusion of patients from the whole country which prevents a falsely too low estimation of incidence caused by treatment of patients in neighboring regions and thus non-inclusion. In the present study, women were affected by BP about 1.2 times more often than men, which corresponds to the results from France and Uganda [3, 13]. In Singapore, women predominated 2 : 1, in Kuwait in a study of patient data from 1991 to 2002 even by 6 : 1 [9, 14]. In these studies – it must be noted – only absolute patient numbers were compared and not the gender-specific incidence. The present analysis, in agreement with Jung et al. , found that in the over 80-year-olds absolutely about double as many women than men were diagnosed with BP, the age-specific incidence of BP due to the small number of older men in the population was, however, significantly higher in men than in women.
Other subepidermal autoimmune bullous disorders
As the second most common autoimmune bullous disorder following BP we identified mucous membrane pemphigoid and pemphigoid gestationis with an incidence of 2.0 new cases/million/year each. As with BP, the incidence of pemphigoid gestationis in the present study is distinctly higher than in France (0.84) and than reported 10 years ago in Lower Franconia (1.0) [2, 3]. In Kuwait, a similar incidence (1.83) was reported . The incidence that we calculated with 1 case to 256 live births was also significantly higher than the observation in the 1970s in England and in the 1980s in Switzerland with 1 case in 3,000 to 4,000 or 7,000 pregnancies, respectively [15, 16]. In agreement with past studies, the women developed the disease usually in the second or third trimester of pregnancy . The distinctly higher incidence in our study might be explained by the fact that we consistently included pemphigoid gestationis in the differential diagnosis in all pregnant patients with long-standing pruritic skin lesions.
Bernard et al. calculated an incidence of mucous membrane pemphigoid in France of 1.25 new cases/million/year , Zillikens et al. an incidence for lower Franconia of 0.67 . For ocular pemphigoid, a mucous membrane pemphigoid with conjunctival involvement exclusively, an incidence of 1 new case/million/year was reported in England . While in studies on large collectives with in part over 400 patients with mucous membrane pemphigoid, a mean age of manifestation was reported between 62 and 66 years [17, 18], the mean age of manifestation of our 4 patients with mucous membrane pemphigoid was 76 years.
The incidence of dermatitis herpetiformis and linear IgA disease was 1 new case/million/year each. The incidence of dermatitis herpetiformis varies regionally very considerably: in Scotland and Sweden prevalences ranging from 115 to 200 patients per million population are reported, in Sweden and Utah, USA, incidences of 11 and 9.8 new cases/million/year, respectively , while the disease appears only sporadically in Japan and Africa . These differences are related to the different immunogenetic backgrounds of the populations. Thus, a clear association of celiac disease with HLA-DQ2 and -DQ8 was found . Dermatitis herpetiformis represents the cutaneous manifestation of celiac disease and practically all patients display at least histological signs of this intestinal disease in jejunal biopsy.
The incidence of linear IgA disease was 1 new case/million/year in our study and thus higher than in Singapore (0.26), Kuwait (0.7) and France (0.48) as well as in Lower Franconia at the beginning of the 1990s (0.22) [2, 3, 9, 14]. Interestingly, linear IgA disease is much more frequent in South Africa, Mali and Uganda [13, 22, 23]. For Uganda for the years 2000 and 2002, it was shown that the autoantibody class correlates with patient age: Younger patients with an autoimmune bullous disorder were more likely to have IgA reactivity against the basement membrane zone and older patients more commonly IgG autoantibodies. The higher proportion of children in African countries (in Uganda 50% of the population is 14 years or younger in comparison to 16% in Germany) thus explains the relative high incidence of linear IgA disease and the relative rarity of BP in Uganda .
Epidermolysis bullosa acquisita belongs with an incidence of 0.5 new cases/million/year in the present study in agreement with surveys in Singapore (0.26), Kuwait (0.23), France (0.19) and Germany (0.22) to the rare autoimmune bullous disorders [2, 3, 9, 14]. With incidences of under 0.3 in France, Singapore and Kuwait bullous systemic lupus erythematosus, lichen planus pemphigoides and anti-p200 pemphigoid are even rarer [3, 9, 14]. During the observation period we did not diagnose these diseases in any patient from Lower Franconia.
In Germany, Finland, France, Bulgaria, Macedonia, Serbia, Turkey, Iran, Connecticut (USA), Saudi Arabia, and Kuwait diseases of the pemphigus group are reported with an incidence of 1 to 6 new cases/million/year [14, 24–27]. Higher incidences possibly exist in Malaysia and China [28, 29]. Recently, an incidence of 7 new cases/million/year was reported in Great Britain . In this study, a registry of all diagnoses in offices of general practitioners was referred to, so that possibly patients with other autoimmune bullous disorders were also included.
In the present study, one patient was diagnosed with pemphigus vulgaris which corresponds to an incidence of 0.5 new cases/million/year. This relatively low incidence compared to Central Europe could be a result of the relative short observation period of 18 months, in which only one patient with this disease was registered. IgA and paraneoplastic pemphigus are so rare that no epidemiologic data have been gathered to date. In the present study, no patients were diagnosed with these two entities or with pemphigus foliaceus.
In summary, this prospective study shows a distinctly higher incidence of the most common autoimmune bullous diseases (BP, pemphigoid gestationis and mucous membrane pemphigoid) in comparison to a similar study in the same region 10 years ago and data for Central Europe up to now. This is probably due to improved diagnostic tools and the increased consciousness of these diseases with the resulting more complete detection of these patients.
We thank Dr. Susanne Herzog, Würzburg, for the help in management of the patients as well as Christa Knaus and Silvana Noll, Würzburg, for the technical support.
Conflicts of interest