Behçet's disease (BD) is a systemic vasculitis of unknown cause, classified among the Variable Vessel Vasculitis by the 2012 Revised International Chapel Hill Consensus Conference.1 It was first described in 1937, by Hulusi Behçet, as a triad of recurrent oral and genital aphthosis and uveitis.2 In addition to these classic symptoms, BD is a chronic disease, with a cumulative course, marked by periods of exacerbations and remissions, that may affect several systems such as articular, vascular, neurological, gastrointestinal, cardiac, pulmonary, or others. Therefore, the disease spectrum ranges from mucocutaneous manifestations to deep organ disease with relevant morbidity and life-threatening involvement.

Although BD has a worldwide distribution, its prevalence is higher along the ancient “silk route” extending from the Mediterranean region to the Middle East and Far East countries.3,4 BD prevalence varies from 20–420/100,000 inhabitants in Turkey to 0.27–0.63/100,000 inhabitants in the United Kingdom.5,6

Organic involvement and BD's clinical course exhibit different phenotypes in different regions, although the factors that lead to these differences are not exact.3,4,7 To date, the etiological factors of BD remains speculative. Nevertheless, infectious agents, genetic causes, immunological susceptibility factors, and fibrinolysis defects have been implicated in BD's etiopathogenesis.8 An association of BD with the presence of HLA-B*51 allele has been reported, which was associated with greater susceptibility to the disease in the countries with the highest prevalence. However, this association does not appear in western countries.9 HLA-B*51 positivity is not related to BD severity.10

The severity of BD manifestations is known to be greater in men due to an increase in morbidity related to ocular, vascular, and neurological manifestations of BD4,10–16 and in younger patients at the onset of disease.4,17–19 It has also been described in cohorts of patients with ocular involvement who present erythema nodosum.11,13

The purpose of this study was to evaluate clinical factors that are associated with the manifestations of severe BD in a single-center cohort.

Our cohort includes 243 patients followed between 1981 and 2020, of which 75 (31%) patients are male. The current age was 49.9 (+14.3) years old (y.o.), the follow-up time was 14.6 (+8.1) years. The first signs and symptoms of the disease started at 28 (+13.3) y.o. B.D diagnosis were established when patients had, at the mean, 34.7 (+12.3) y.o. BD diagnosis was established, on average, 6.6 (+7.8) years after the onset of the first symptoms, and 0.6 (+2.2) years after admission to our consultation.

The HLA-B*51 allele was only accessed in 200 (82%) patients and present in 108 (54%). The mean Krause CSS was 5.7 (+3.5) points and a BODI score of 0.7 (+1.5) points.

Regarding organ manifestations, during the follow-up period, all patients had mucosae manifestations (N=243, 100%), 133 (55%) had cutaneous, 104 (43%) joint, 71 (29%) ocular, 48 (20%) vascular, 47 (19%) neurological, 22 (9%) gastrointestinal and 1 (0.4%) cardiac involvement by BD. Manifestations by organ are detailed in Table 1.

In most patients, mucosae manifestations were the first symptoms of BD (N=234, 96%). Skin involvement was present at BD onset in 29 (12%) patients. Patients exhibited signs of skin involvement seven years after the appearance of first signs of BD. Articular, ocular, vascular, gastrointestinal, and cardiac manifestations appear later in the disease course, more than nine years after the first BD signs. Table 2 presents the time elapsed between first signs of BD and each organ manifestation.

Besides BD, 22 patients presented overlap with other autoimmune diseases, namely, seven patients with Sjögren's syndrome, five with ankylosing spondylitis, two with psoriatic arthritis, two with psoriasis without joint involvement, two with antiphospholipid syndrome, one with rheumatoid arthritis, one with idiopathic thrombocytopenic purpura and another with temporal arteritis.

Although associations between clinical manifestations of BD and a more severe course of the disease have been described, most of these associations were described in subgroups of patients who have a specific type of organ involvement, such as ocular, intestinal, among others.4,10–13,15–19,25 Our cohort included all patients followed at our center, regardless of specific organ involvement. This study revealed the association of the male gender as well as the presence of erythema nodosum and pseudofolliculitis, as clinical characteristics that are independently associated with manifestations of severe BD.

As for gender differences, several studies reported a higher incidence of ocular, vascular, and neurological impairment in men,10,12,15,16 although in the latter case, the results were sometimes contradictory.14 Using a modified form of the BD total activity index to access BD severity, Gül10 in a cohort of 148 patients, described an increased risk of severe disease in men, in a logistic regression including demographic data, family history, and HLA-B*51 allele positivity, without including other clinical features in the analysis. Furthermore, Balta12 described a higher incidence of ocular, neurological, and vascular manifestations in men, comparing each organic manifestation's prevalence between men and women. Likewise, Cansu15 described a higher incidence of vascular and ocular damage in men, just as Davatchi16 described a strong association between male sex and vascular damage. In a study that addresses BD's mortality conducted by Saadoun,26 male sex and arterial involvement were independently associated with mortality. On the other hand, Bang14 reported a higher prevalence of neurological disorders in women compared to men. Based on these studies, it is known that the disease has a more severe phenotype in men. However, the variability in the BD severity definition used in different studies makes it difficult to compare results. Our study corroborates this association, demonstrating the independent association of male gender with the most severe forms of BD in the global expression of the disease and not only limited to a subgroup of clinical manifestations.

The presence of erythema nodosum as a prognostic factor for the severe disease was only described in patients with ocular disease. Likewise, the definitions used to address BD's severity in these ocular cohorts differ. Using a worse visual prognosis as an outcome in a cohort of ocular BD, Hu11 described a more severe BD in males presenting erythema nodosum. On the other hand, defining severe disease as death, blindness, vascular, neurological, lung, gastrointestinal and articular BD involvement, Zouboulis13 described a more severe disease in patients presenting a triad of symptoms: ocular BD, erythema nodosum, and thrombophlebitis. Our study shows that erythema nodosum is independently associated with the most severe manifestations of BD and that this association is not limited to ocular BD cohorts.

A higher incidence of pseudofolliculitis has been more described in males than in females in several studies,12,18,19 although others found no differences between genders.15 The presence of pseudofolliculitis had not previously been described as a factor independently associated with severe BD manifestations. The prevalence of pseudofolliculitis is very variable in different geographical locations, with a reported prevalence of 39% from Egypt to 91% in Jordan.7 In our population, the prevalence of pseudofolliculitis is low (39.5%) compared to other series with many patients included.25,27,28 However, given BD's phenotypic variation in different locations, this association described in our cohort may not be reproducible in other populations. Further studies extended to other population groups are necessary to confirm it.

In our study, the increase of the current age is associated with severe BD, with the odds for severe BD increasing 2% per year, without considering interaction with other factors. This finding is in accordance with the disease's progressive course and the appearance of organ manifestations over the years. However, when age was analyzed together with other factors, we found no association between severe BD and the patient's age. The onset of BD clinical manifestations is around the third decade of life in most studies,7 the same in our population, 28.1 years. We did not find any association between the age of onset of symptoms and severe BD. Studies that reported this association analyzed the effect of age without considering interaction with other factors18,19 or considered only a group of patients with a specific organ target.17,29,30 Yazici18 demonstrated that men with the onset of symptoms before 24 y.o. had more eye disease. Alpsoy19 showed that patients with age at onset under 40 y.o. had more ocular involvement and genital ulcers. In patients with intestinal BD, Jung17 reported that age at onset, greater severity of the intestinal disease, and the prevalence of volcano-shaped ulcers are associated with a worse prognosis. Also, in patients with intestinal BD, Zhang30 associated the earlier age at onset with shorter event-free survival. Similarly, in a cohort with ocular BD, Belkhadir29 describes a worse ocular outcome in patients with earlier age at onset and with a higher number of recurrences. The authors raise the hypothesis that this association with age may be due to greater exposure to BD flares throughout life, with periods of exacerbation and cumulative damage and not precisely more severe manifestations when they appear at an earlier age, however, this study does not allow us to draw these connections.

As in previous studies, there was no association between the HLA-B*51 allele's positivity and the presence of severe BD.10

Regarding the sequence of manifestations’ appearance, in most patients of our cohort, mucosae manifestations were the first to appear, followed by skin involvement, around seven years later. Other organ manifestations arose almost a decade later. Thus, we believe that these clinical characteristics can assist the clinician in the follow-up of these patients and serve as a warning sign for earlier detection and timely treatment and prevention of deep organ damage. It is important to notice in these time interval's analysis, that the real differences may not be that big. On the one hand, both mucosal and cutaneous manifestations are detectable without the use of complementary exams and, therefore, they can either be either noticed by the patient or be observed by the clinic, and thus, be detected practically at the onset time. On the other hand, usually deep organ involvement diagnosis requires complementary tests and, therefore, can be detected later. Over the past few years, technological developments in terms of complementary tests have increased the detection of BD's deep organ manifestations, namely vascular and neurological involvement, among others. In our study, due to the long follow-up period, these manifestations, especially for the patients included within the first years, may be underrepresented, although the long inclusion period together with the cumulative BD manifestations presented, tends to lessen this problem.

BD diagnosis relies basically on the identification of several clinical characteristics, that can emerge separated by several years, enough to fulfill the classification criteria (with a specificity above 95% 31).32,33 This can represent underdiagnosis in the initial stages and accounts for the lag time between the BD onset and diagnosis, on average, 7 years in our study. Furthermore, some BD phenotypes can mimic other autoinflammatory diseases, especially the antineutrophil cytoplasmatic antibodies (ANCA)-associated vasculitis (AAV).34 The distinction between these entities can be a challenge, due to manifestations overlap. ANCA positivity does not always help in the diagnosis: there are ANCA-negative AAV, and, on the other hand, ANCA has been detected in a wide range of diseases, other than AAV35, including BD with vascular involvement.34 To date, no known biomarker has been reported for BD's, but efforts have been made.32 Both factors can introduce a diagnostic bias in our study, underrepresenting BD cases at an earlier stage and, also, patients with atypical presentations. The authors believe that, in the future, the identification of such biomarkers can help to distinguish these diseases, and possibly allow an earlier diagnosis.

Our study's main strengths lie in the fact that we have considered the multiorgan manifestations of BD and were not limited to a specific organ, in the high number of patients followed in our center, and in the long follow-up period. The main weaknesses of this study are due to its observational nature with its inherent limitations as well as to the fact that we used a definition for severe BD a score with a final result that can translate just one or an association of organs, which can hide whether these associations are due to a preferential organ achievement. In our opinion, our study raises future questions, namely whether these associations vary with gender or with the involvement of a specific organ.

Male gender, erythema nodosum, and pseudofolliculitis were associated with severe DB. Mucous manifestations are usually the first sign of the disease and, together with skin involvement, usually, appear before other organs are affected. Thus, these manifestations can help as an alert for the clinician, indicating that the patient who presents them may develop severe BD manifestations in the future.

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

None.