Mucocutaneous Findings in Behçet’s Disease

Behçet’s disease is a chronic inflammatory disease with an unpredictable course. The dis - ease may affect almost all organ systems resulting with significant organ-threatening morbidity and mortality. Mucocutaneous lesions mostly constitute the initial symptoms of the disease and precede other manifestations. As there is yet no pathognomonic diag nostic test in Behçet’s disease, the recognition of cutaneous and mucosal findings let the physician enable an earlier diagnosis and earlier treatment. Therefore, the purpose of this chapter is to emphasize the importance of the mucocutaneous manifestations of Behçet’s disease and to review the mucocutaneous lesions in detail. Finally, childhood Behçet’s disease, differential diagnosis and treatment of mucocutaneous manifestations will be briefly reviewed.


Introduction
Behçet's disease (BD) is a chronic, relapsing inflammatory multi-systemic disease of unknown etiology with a course of exacerbations and remissions [1][2][3][4]. Prevalence of BD is higher in countries lying along the ancient Silk Road, extending from eastern Mediterranean to East Asia [5]. Turkey probably has the highest occurrence level of the disease with prevalences of 110-420 cases per 100,000 population [6][7][8]. Today, due to immigrations, BD is encountered almost all over the world [9,10].
The disease was first defined in 1937 by the Turkish dermatologist named as 'Hulusi Behçet' with the presence of 'triple symptom complex' of recurrent oral ulcers (OU), genital ulcers (GU) and uveitis [1]. After the initial description, it now became increasingly evident that BD is a multi-systemic disease with involvement of mucocutaneous, vascular, neurological, musculoskeletal and gastrointestinal systems with significant morbidity and mortality [2,3,9,11,12]. The main histopathological finding is the vasculitis of the arteries and veins of any size or thrombophilia according to the site of involvement [13].
Although the exact etiopathogenesis of BD is still unknown, it has been hypothesized that in genetically predisposed individuals, a development of an inflammatory reaction against to an infectious, an environmental or an autoantigen and/or the presence of disturbances in molecular mechanisms in regulating immune responses may contribute to the disease [2,3,11,[14][15][16][17].
Today, most of the authors classify BD as a group of systemic vasculitis (under the title of variable vessel vasculitis) as a result of the consensus; '2012 Revised International Chapel Hill Consensus Conference Nomenclature of Vasculitides' [18].
The first symptoms of BD usually occur in the third and fourth decade of life [2,4,9,19]. However, childhood cases have also been reported [20]. Juvenile BD was found in 7.7% and family history was found in 11.6% of the patients in the study reported by Alpsoy et al. [21]. In another study by Gurler et al., the presence of family history in the first degree relatives of patients with BD has been reported as 7.3% [22]. Male patients, a younger onset and HLA-B51 positive patients are found to have more severe kind of the disease [11,21,23,24]. Contrary to these results, Davatchi et al. found no association between severe organ involvement and male gender except vascular involvement [25]. Clinical features of BD include OU, GU, ocular inflammation, cutaneous lesions, as well as articular, vascular, neurological, pulmonary, gastrointestinal, renal and genitourinary manifestations [2][3][4]11]. BD may start with just one or two symptoms but other symptoms may gradually appear over the years [19,22,24,[26][27][28][29].
As there is no pathognomonic test, the diagnosis of BD depends on clinical criteria [3]. In 1990, the International Study Group (ISG) for BD defined new diagnostic criteria by reviewing the data of 914 patients from 12 centres in seven countries around the world. The ISG criteria for BD have a sensitivity of 92% and specificity of 97%, compared with previous sets of criteria. According to these criteria, the diagnosis of BD consists of the presence of recurrent OU in addition to two of the following features: GU, eye involvement, skin lesions and positive skin pathergy test. At minimum three episodes of oral aphthous ulcer in a year should be observed for the diagnosis of BD [30]. In order to increase the sensitivity and specificity, these criteria are re-assessed and revised [31]. Since patients may present with only OU for a long time showing a long prediagnostic duration [19,22,32].
This chapter aims not only to define the mucocutaneous lesions but also to elaborate the features of the mucosal and cutaneous lesions in detail. Furthermore, mucocutaneous lesions in paediatric BD, differential diagnoses and the management of the mucocutaneous lesions will briefly be reviewed.
The above-mentioned mucocutaneous lesions may be the initial findings of BD or may be observed at any time during the course of the disease [21,22,24,28,29]. Especially, OU is the most commonly detected lesion and mostly emerges as the initial clinical finding of the disease [2,3,19,21,[33][34][35][36][37]. OU and GU are mostly considered as the 'fingerprint lesions'. As there is yet no pathognomonic test for the diagnosis of BD, recognition of these mucocutaneous lesions let the clinician make an earlier diagnosis, which also enables an earlier treatment [2,3].
OUs are recurrent and painful ulcerations of the oral mucosa characterized by a round or oval ulceration with sharp borders surrounded by a red erythematous inflammatory area (Figure 1). The base of the OU is necrotic with a yellowish-white colour. The most common sites are the mucous membranes of the lips, buccal mucosae, tongue, uvula and soft palate [2,3,22]. Clinically, they look like similar in appearance with conventional aphthae that may be seen in several systemic diseases such as inflammatory bowel disease, Sweet syndrome, systemic lupus erythematosus or recurrent aphthous stomatitis (RAS) [2,12,32]. OU in BD tends to recur more frequently and to be more in number [38]. Main and Chamberlain reported that OUs in BD have a more diffuse erythematous surrounding rim, they are localized more often in soft palate and oropharynx and they are more in number when compared with the aphthae of RAS [39]. However, no difference was found in terms of frequency of OU between the two groups [39].
OUs are usually classified into three groups based on the diameter of the ulcer [2, 11]: 1. Minor aphthae are shallow mucosal ulcers with a diameter less than 10 mm. It usually spontaneously regresses in a couple of weeks without formation of scarring.

2.
Major aphthae have a deeper morphology with a diameter larger than 10 mm. Healing usually occurs in 3-4 weeks with scarring.

3.
Herpetiform aphthae are pinpoint shaped, very small and shallow mucosal ulcers and localized in crops.
The majority of the patients were found to have minor aphthous ulcer (75%) followed by herpetiform (20%) and major aphthae (5%) in the study of Vaiopoulos et al. [36]. Different sizes and types of OU may be seen at the same time in oral mucosa in BD [9]. A study comparing the characteristics of OU between BD and RAS was performed by Oh et al. They reported that major OUs were significantly more common in patients with BD, and initiation of OU in BD was more likely related to menstrual cycle when compared with OU in patients with RAS.
In addition, they also concluded that patients with major aphthae who have accompanying articular symptoms as initial symptoms should be strictly followed for the possible development of BD [40].
Ideguchi et al. evaluated 412 BD patients' data including 16 years follow-up. The results revealed that a mean of 7.5 years has been proceeded before a definitive diagnosis of BD. In the same study, 14% of the patients had suffered from OU for more than 20 years before the diagnosis of BD [19].
In two different studies reported by Alpsoy et al., the mean duration between the OU and the fulfilment of diagnostic criteria was determined to be 4.3 ± 5.7 and 3.77 ± 4.43 years, respectively, and it has been concluded that the disease is often diagnosed with a delay of several years after the appearance of the first sign [21,26].
Although OUs are the most common and earliest manifestation of BD, a few studies have been reported indicating a course without the presence of OU at the onset of BD [27,[41][42][43][44].
On the basis of this issue, Faezi et al. evaluated the clinical features on 175 patients with BD who do not have oral aphthosis (NOA) and compared them with the patients with OU. The results revealed that the first manifestation was uveitis with a ratio of 70.3% in the NOA group, and a pathergy test was more common in NOA group [41].

Genital ulcers
Genital ulcers (GUs) are generally the second most frequent disease manifestation [2,9,21,26,28,45,46]. GUs are observed in approximately 60-97% of the patients with BD at any time of the course of the disease [4,19,21,22,28,36,43,47]. The lowest frequency was reported from Romania with a ratio of 55.5% [43]. As an initial symptom, GU was found in 14.2 and 7.4% of the patients with BD in two different studies from Turkey [21,22] while in only 4% of the patients with BD in a study from Moscow [34].
GUs are usually localized on the scrotum and on the shaft of the penis in male patients while on the labium major and minor in females (Figures 2 and 3). They are rarely seen in the vagina and cervix in females and on the urethral orifice and glans penis in males [2,3,9,11,22]. They are usually painful or very rarely they can be asymptomatic [2]. They have mostly the same characteristic morphological features of OU. The clinical differences from oral aphthous ulcer include that GUs are larger and deeper, have more irregular borders and a longer healing duration. In addition, GU recurs less often and heal with scarring [3,32,48]. Therefore, in the suspicion of diagnosis of BD, scatris of GU on genital region should also be searched.
While some studies have documented higher frequency of GU among one of the genders [21,24,35], some did not detect any difference between the two genders [23,25,34]. Gender differences (Male>female) at the onset of GU have been observed in the study by Vaiopoulos et al. [36].
GU may cause severe pain, difficulty in micturition, dyspareunia and marked difficulty in physical activity. Deep ulcers located in vagina may be complicated by fistulisation to bladder, urethra or rectum [2][3][4]9].
Mat et al. reported a prospective study investigating the frequency of scarring after GU. This study revealed that healing with scarring was observed in 49% of small GU while in 89% of large ulcers. They stated that ulcers on labium minors and vestibule may heal even without scar formation [48].
Faezi et al. reported 64.7% of 6935 cases with BD had GU during the course of the disease. They compared clinical and laboratory features between the patients with GU and patients who never developed GU (non-GU cases). As a result of their study, OU and other cutaneous manifestations such as pseudofolliculitis and erythema nodosum were found to be higher while eye involvement was found to be less common in the GU group [46].
PPLs are usually localized on the trunk, buttocks and extremities [2,3,49]. However, they may be observed on any part of the body and even on palmoplantar regions [9]. In case of face and chest localization, especially in adolescence period, it may be hard to distinguish PPL from an ordinary papul/pustule of acne or folliculitis [2].
PPLs were found to occur more frequently in male patients than females in a study by Tursen et al. [35]. PPLs were found to occur more in patients with positive pathergy test [49].
Since PPLs are non-specific and resemble to ordinary acne pustules and/or folliculitis; some authors suggest that non-follicular lesions located other than face are more valuable for the diagnosis of BD [2,27,49] while some suggest that PPL should not be included as a diagnostic criteria [50]. It is suggested to accept PPL as a diagnostic criteria only if leukocytoclastic vasculitis or a neutrophilic reaction histopathologically is detected [2,50].

Erythema nodosum-like lesions
Erythema nodosum (EN)-like lesions are frequently observed skin lesions in patients with BD [9]. The incidences of EN-like lesions have been reported between 15 and 60% in various studies [21,22,24,28,55,56]. In 2.8% of the patients with BD, EN-like lesions have been reported as an initial symptom [22]. They are more common in female patients [23,24,36,55]. They are characterized by multiple painful subcutaneous nodules with different sizes. Although they are preferentially located on the lower limbs, they may also be localized on gluteal region, upper extremities and neck [2,3,9,22,55]. They heal within 10-20 days without secondary ulceration and scatris formation. The resolution generally results with residual hyperpigmentation [2,3].
Superficial TFB is mostly characterized by linearly arranged erythematous subcutaneous nodules in lower extremities that migrate from day to day. The vein can be palpated as a string-like hardening showing the thrombosis and vessel sclerosis [2,3]. EN-like lesions and migratory TFB may be thought as the differential diagnosis [2,3]. Superficial TFB is clinically important since it is frequently associated with other forms of vascular disease in BD [2,63].

Extragenital ulcers
Extragenital ulcer (EGU) is usually a solitary, small, round ulcer with a red rim and yellow base; however, it may have various shapes and sizes [2,3,32,65]. Although observed rarely, it is the most characteristic and specific lesions of BD [2,65]. EGU may be localized anywhere on the body such as legs, axillae, breast, interdigital skin of the foot, neck and inguinal regions. EGU may persist for a long duration, may be painful and heals usually with scarring [2,3]. In a report of Azizlerli et al., four cases of EGU were shown to reveal vasculitis histopathologically [65].
A study reported by Ozyurt et al. detected that the patients with family history of BD had more frequent EGU than patients with negative family history of BD [66].
Few series documented common EGU in children [67,68]. However, due to rarity of BD in childhood and few reports concerning the clinical findings in children with BD, the characteristics of the disease in this age group are not completely described [67][68][69].

Sweet's syndrome-like lesions
Few cases of Sweet's syndrome-like lesions have been reported in patients with BD [70][71][72][73][74][75][76][77]. Sweet's syndrome-like lesions are characterized by painful, edematous papules, plaques and nodules localized on face, neck and back [3,[70][71][72][73][74][75]. Fever and laboratory findings such as leukocytosis, elevated erythrocyte sedimentation rate and C-reactive protein accompany to the cutaneous lesions [73,74]. Although clinical and histological overlap exists between Sweet's syndrome and BD such as the presence of OU, arthralgia, arthritis, episcleritis, pathergy positivity and neutrophilic infiltrate in the dermis in both of the diseases, there are some distinguishing features. In BD, the development of OU is more frequent, fever is rarely seen, the pattern of articular and ocular involvement is different [3]. In addition, comparing two diseases by human leucocyte antigen (HLA) typing revealed that patients with BD had higher frequencies of HLA-B51 and HLA-Dqw3, while patients with Sweet's syndrome had higher frequencies of HLA-Bw4 [77]. Sweet's syndrome-like lesions have been reported to occur in the acute phase of BD or sometimes have been thought to point a flare in BD [73].

Pyoderma gangrenosum-like lesions
Pyoderma gangrenosum (PG) is another rare neutrophilic dermatosis that may be associated with BD [78,79]. Rare cases of PG-like lesions in patients with BD have been addressed [78][79][80][81][82][83][84][85]. The lesion is usually characterized by large superficial ulceration localized usually on the buttock or the lower limbs; however, it may reveal vegetative or bullous variants [9,79,81]. Both diseases may have clinical and histopathological overlap. The patients with PG may also have OU, GU and pathergy positivity [78,79]. However, as mentioned above, the frequency of OU and GU is higher in BD. It has been reported that PG is associated with the activation of BD [78,79,85]. Also, Hali et al. reported two paediatric cases of BD and PG with a fatal outcome [84].
Although it is a rare finding, it has been suggested to be a specific clue for establishing the diagnosis [102,103].

Skin pathergy test
Skin pathergy test (SPT) is a non-specific hyperreactive reaction of the skin that occurs as a response to a minor trauma such as a needle prick [2,3]. SPT was first described by Blobner in 1937 [70,105]. It is used as an adjunctive test in the diagnosis of BD and according to the ISG criteria for BD, a positive SPT is a criteria needed for the diagnosis of BD [2,3,30,70,105].
Positive SPT is defined as the development of erythematous, indurated papule which usually evolves into a sterile pustule at the site of the needle puncture after 24-48 hours (Figure 4) [70,105]. Although the exact mechanism of the SPT is still not known, it is thought to occur as a result of enhanced non-specific inflammatory response and aberrant release of cytokines triggered by the cutaneous injury [13]. A number of studies investigating the histopathological examinations of positive SPT reaction (papule/pustule) revealed findings ranging from mononuclear cell infiltration at varying densities in perivascular or periadnexal areas to leukocytoclastic vasculitis [106][107][108][109][110]. Ergun et al. evaluated a chronological histopathological study of sites of SPT and observed intraepidermal pustules and polymorphonuclear infiltrate at the beginning of the inflammation [108]. Androjen receptor levels were found higher in positive SPT sites when compared with normal skin [111].
Although the positivity of SPT has been accepted as a criteria in the diagnosis of BD, there is no consensus about performing a standardized method of SPT [105,[112][113][114][115][116][117]. It is usually performed under sterile conditions with a 20-gauge needle inserted intradermally into the avascular area on the forearm skin of patient with an angle of 45° [105,112,113]. Various techniques such as pricking with multiple needles with various applying routes including intradermal, intravenous and subcutaneous methods have been reported [9,105,112,[114][115][116]. Multiple punctures are mostly required [9,114,115]. Davatchi et al. suggested three intradermal punctures perpendicular or diagonally one with a 25-gauge needle with intradermal injection of one drop of normal serum saline, one with a 25-gauge needle alone (just a puncture, with no injection) and the last with a 21 gauge needle (puncture, no injection) [115]. A study by Ozdemir et al. suggested that two needle pricks are sufficient for positive SPT [114]. Another study by Ozdemir et al. analysed the changes of SPT positivity in different body areas such as flexor surfaces of the forearms, the lateral aspect of the tibial area, the scapular areas on back, and the lumbar areas of the abdominal region. They concluded that forearm was the most frequent site positive for pathergy reaction whereas abdomen was the least [113]. Akmaz et al. reported higher rate of positivity in SPT by intradermal application compared with intravenous application [117]. Dilsen et al. performed SPT with different needles including sharp and blunt needles in which they confirmed higher frequency of positivity with blunt needles [112]. Sharquie et al. demonstrated an alternative method of which they inserted the needle inside the mucous membrane of the lower lip to the sub-mucosa. However, the sensitivity of the oral pathergy test has been reported lower than of the classical pathergy test [118].
Various prevalence rates of SPT positivity in BD have been reported by several studies [21,22,[119][120][121][122][123]. The positivity of SPT varies between 40 and 88% with a higher prevalence in Japan and Mediterranean countries, whereas it is lower in countries such as the United Kingdom and the United States [2,3,9,22,45,98,[119][120][121][122][123]. Alpsoy et al. reported the positivity of SPT as 37.8% in their study including 661 Turkish patients with BD [21]. A study comparing the positivity of SPT between Turkish and British patients with BD revealed the positive reaction was only present among Turkish patients [119]. In a German study, SPT revealed positive results in 33.7% of patients with no significant difference between Turkish and German patients with BD [123]. The factors that may affect the rates of positivity of SPT include genetic variables, variations in ethnic origins of the patients, factors related to the method and materials used to perform the SPT and conditions of the patient and the disease [105,[112][113][114][124][125][126].
Contrary results have been reported about the relationship between positive SPT and clinical course of BD [22,[125][126][127][128]. Chang et al. found that positive SPT was specific for BD, but not associated with clinical severity [125]. It has been reported that the frequency of SPT positivity has decreased during the last years [115,125,129]. One of the best reasons for this issue is the use of disposable needles, which are less traumatic than non-disposable ones [105,112,125]. A study by Davatchi et al. investigated the sensitivity and specificity of SPT among 6607 patients followed between the years 1975 and 2010. Their results revealed that the sensitivity of pathergy test has declined, but it still preserves its specificity [115]. Moreover, positivity of SPT in many diseases other than BD has been detected. Neutrophilic dermatoses such as Sweet syndrome, PG, erythema elevatum or other diseases such as RAS, eosinophilic pustular folliculitis, inflammatory bowel diseases and spondyloarthropathies are some in which pathergy reaction positivity has been shown [70]. Despite everything, SPT has still a diagnostic value [115].
The diagnosis of paediatric BD in the reported studies was based on the criteria of ISG and ICD [30,31]. A recent classification was proposed by Kone-Paut et al., and according to this classification, the presence of any three items among recurrent oral aphthosis, GU, ocular involvement, neurological and vascular signs makes the definitive diagnosis of BD [135,140].
The onset of BD has been reported at any age between 0 and 16 years [131,136,137]. The lowest mean age of 4.8 years old was reported by Nanthapisal et al. [134]. Similarly to clinical symptoms of adulthood BD, OU was present in most of the children and also constituted the initial symptom in most of the cases [68,[131][132][133][134][135][136][137][138][139]. The localization and morphology of OU in children were also similar to the characteristics of adulthood OU [136,137,140]. GU was detected between 55 and 94% of the paediatric cases [131,[135][136][137][138]. Distinctively from adult cases, the presence of GU has been reported significantly lesser than adults [131,132]. However, Treudler et al. did not reveal any differences in the ratio of GU between childhood and adulthood BD [68], and Nanthapisal et al. detected recurrent GU more frequently in females [134]. The characteristics of GU were also found similar to the ones seen in adults [136,137]. Another distinctive feature, perianal aphthosis was to be a specific feature of childhood BD [135].
Skin lesions were reported to occur between 64 and 92% of childhood BD [70,135,137,138].  [136]. However, no gender differences were detected in terms of skin lesions in the study of Borlu et al. [137]. Erythema multiforme-like lesions and TFB have been less frequently reported [70]. Positive SPT has been reported in 37, 80 and 76% of the patients in three different studies [131,136,137].
Nanthapisal et al. documented skin findings in 11 (23.9%) of 46 patients: These were pustular lesions in three, skin ulceration in two, EN in two, necrotic folliculitis in two, PPL in two and positive SPT in three of the five patients [134].
The duration between the initial symptom and fulfilment of diagnosis of BD has been reported as 3.14 years in the study of Karincaoglu et al. [131], while a significant diagnostic delay up to 13.5 years has been reported in the study of Nanthapisal et al. [134]. This may be caused due to rarity and unfamiliarity of the disease in Northern Europe.
Conflicting reports have also been found in the systemic expression of BD in children [130,132,133]. BD was usually reported to have a less severe disease [132,140]. Pivetti-Pezzi et al. reported similar rates of OU, GU, skin lesions in the comparison of adulthood and childhood BD. In addition, no differences were observed in the incidence of arthritis, gastrointestinal and neurological involvement except more severe ocular involvement in childhood. Another study by Sarica et al. compared the patients with mild and severe diseases. An earlier onset and more systemic involvement were found in patients with severe form of the disease [130]. More frequent neurological and gastrointestinal involvements were observed in childhood BD in the study of Karincaoglu et al. [131].
Generally, it can be interpreted that juvenile BD has a similar clinical spectrum to adulthood BD. The differences in frequency and clinical courses may reflect the geographic, ethnic and genetic variations.

Differential diagnosis of mucocutaneous manifestations
The differential diagnosis of mucocutaneous manifestations will not be addressed in detail as this subject does not constitute the main topic of this chapter. Only the essential differential diagnoses of OU, GU, EN-like lesions will be given below.
In case of more severe and painful OU and/or GU, a short duration of systemic steroids may be added with colchicine. Corticosteroids combined with systemic antibiotics can be used to decrease the severity of GU attacks [147][148][149]. In patients with severe mucocutaneous manifestations, immunosuppressive drugs such as azathioprine (AZA), methotrexate (Mtx), cyclosporine A may be used [149]. AZA has been found effective in preventing the recurrences of thrombophlebitis [147]. Thalidomide is another choice of therapy in recalcitrant OU and/or GU. However, it is not preferred due its high toxic effects [147,149,150]. Pentoxifylline, dapson, zinc sulphate, IFN-α and rebamipide are other alternative treatments worth for trying in OU. However, larger and well organized studies are needed in order to clarify their efficacies. In case of EN-like lesions, bed rest is usually required. Especially in female cases with GU and EN-like lesions, the combination of colchicine and benzathine penicillin is recommended [148]. This treatment combination has also been reported to decrease the frequency and duration of both OU and EN-like lesions [150]. IFN-α has been reported to decrease not only the frequency of GU and EN-like lesions but also the number of PPL [151]. Anti-TNF drugs are being used with success in patients with refractory mucocutaneous manifestations [151,152]. Recent studies of interleukin-1 (IL-1) inhibitors (anakinra, canakinumab) have demonstrated efficacy in OU and GU resistant to conventional therapy [150]. Finally, apremilast, phosphodiesterase 4 inhibitor, has been reported to be effective in treating OU and GU [150,152]. The treatment options can be seen more detailed in Table 1.

Conclusion
Mucocutaneous lesions are the most important criteria in establishing the diagnosis of BD. In case of recurrent OU, GU and other cutaneous findings mentioned above, it is important to remind the possibility of BD in the diagnosis, which will permit an earlier diagnosis and enable a decrease in mortality and morbidity.

Mucocutaneous manifestation Topical Systemic
Oral