Behçet’s disease (BD) or Adamantiades-Behcet's Disease is a chronic, relapsing, vasculitis of unknown aetiology characterized by mucocutaneous, ocular, articular, vascular, gastrointestinal, and central nervous system manifestations [
1]. BD significantly increases morbidity and mortality. BD mainly affects young adults with a peculiar geographical distribution, also named the “silk-road” – corresponding to the ancient route between the Mediterranean, the Middle East and the Far East. With the exception of oral aphthosis, BD is characterized by considerable phenotypic variation. Over the last 30 years, a substantial body of knowledge has accumulated supporting a strong genetic underpinning in BD of the MHC-related allele HLA-B5, which was later more specifically linked to its predominant suballele HLA-B51 [
2]. It has been suggested that the disease varies in its phenotypic expression in different races and in different countries. Both environmental and genetic factors play a role in the aetiology of the condition [
3]. For instance, BD patients from Asia exhibit a higher frequency of gastrointestinal involvement compared to those from the Mediterranean basin [
4]. It has been reported that there is a greater risk of ocular involvement in patients in Japan or Iran and a lower risk of genital ulceration in most non-western countries [
3]. However, most of the evidence supporting these propositions arises from observational case series, which are subject of many sources of bias. Moreover, to our knowledge no data are available regarding BD in sub Saharan African population. The present study investigated potential ethnicity-related differences relative to phenotype and prognosis of BD patients in a French multiethnic country. To this aim, we compared the main features of BD, the outcome and the factors associated with mortality in the 3 largest ethnic groups of our cohort (i.e. European, North African and sub Saharan African patients).
Discussion
Ethnic origin is one of the factors that may modulate the prevalence and expression of BD. Analysis of the literature showed that studies assessing ethnicity related differences in BD are scarce and often derived from small samples thus preventing clear conclusions. Herein, we reported ethnicity-related differences with respect to phenotype and prognosis of BD in a French multiethnic country.
The comparison of our BD patients according to their ethnic origin showed a higher frequency of male in sub Saharan and North African patients as compared to those from Europe. Similar to variations in the clinical manifestations, gender distribution in BD patients varies widely depends on their ethnic origin and country of residence. For instance, the percentage of male BD patients in a recent study ranges from 27% in USA to 87% in Azerbaijan [
17]. Male gender is a main factor associated with mortality in BD [
18,
19]. In our previous study, 92.7% of the BD patients who died were of male gender [
19]. In multivariate analysis, male gender increased by 5 times the odds of death. Male gender and a younger age at onset have been previously reported to markedly influenced disease expression and course of BD [
18]. Lastly, male patients tended to have more flare of BD compared to female [
19].
We observed ethnicity-related differences regarding phenotype of BD. Our study shows a 1.6 times higher frequency of CNS involvement among patients from sub Saharan Africa as compared to those from North Africa and Europe. Cardiovacular involvement tended to be more frequent in sub-Saharan African patients but difference did not reach statistical significance. Arterial and cardiac complications are less common than venous lesions in BD, occurring in 1 to 7% of patients [
20]. The concept of vasculo-Behçet’s has been adopted for cases in which vascular complications are present and often dominate the clinical feature that fit with the phenotype of our sub Saharan African BD patients. The main causes of death in BD included major vessel disease, and central nervous system involvement [
18,
19]. The frequency of CNS involvement in BD varies widely according country and ranges from 3% in Iran to 34% in Saudi Arabia [
17]. The 48% rate of CNS lesions found in our sub Saharan African BD patients was clearly higher than the frequency observed in Saudi Arabia. Although not independently associated with mortality, CNS involvement accounted for 12% of deaths in our previous study [
19]. In large studies addressing neurologic disease of BD, the mortality rate range between 5.5 and 20% [
21,
22]. The median period until death was of 4 years after neurological onset [
21]. It was previously reported that BD has diverse clinical expression in various geographical areas. The pathergy reaction is considered highly sensitive and specific for BD in patients from Turkey and Japan, yet is frequently negative in patients from Western countries [
23], or gastrointestinal (GI) involvement, which occurs in about one-third of patients from Japan, but rarely in Mediterranean countries. O’Neill et al. [
24], described regional differences regarding several clinical manifestations of BD. They reported that BD patients from Middle Eastern countries and the Mediterranean basin generally have less widespread disease compared with patients from Western countries (i.e. UK and USA), manifested by lower rates of arthritis, vascular problems, and CNS abnormalities [
24].
The HLA B5 allele was two times less frequently found in sub Saharan African BD patients as compared to those from Europe and North Africa and with the overall prevalence of HLA B5 allele usually reported in BD studies [
25]. Large association analysis studies have reported that HLA B51 is carried by one-to two-thirds of BD patients and increases the risk of BD development by 6 [
25]. Authors have suggested that HLA B51 positive and negative BD patients differed in that the former more frequently developed eye involvement [
26] or sometimes CNS involvement [
1] and the latter more commonly thrombophlebitis. However, based on two recent large meta-analyses, it has been shown that no real association exists between HLA-B51 positivity and the frequency of CNS involvement [
25]. In contrast, uveitis, genital ulcerations and skin lesions were more frequent in BD patients carrying this allele. However, whether BD carriers of HLA-B51 allele exhibit a more severe disease course is still unknown.
The most original finding of our study is that sub Saharan African BD patients had up to 3 times higher mortality compared to North African and European patients. Their 15-year mortality rate was of 20% which was much higher than the 5-10% usually found in large series of BD patients [
18,
19]. The sub Saharan African group was independently associated with a poor outcome as well as male gender and cardiovascular involvement in our BD patients. These data has, to the best of our knowledge, never been described in the literature. We suggest that a particular attention should be given to sub Saharan African patients in BD. Krause
et al. [
8] have described the influence of ethnic origin on clinical expression and disease severity in Israeli patients. They studied 100 patients fulfilling International Study Group criteria for BD including 66 Jewish and 34 Arabic patients. The 3 largest ethnic groups of Jewish patients were from Iran/Iraq (n = 21), Turkey (n = 12), and North Africa (n = 21) countries. Arabic patients had more severe ocular disease with significantly higher rate of posterior uveitis (20.6 vs 4.6%). In the 3 most common Jewish ethnic groups, patients form Iran/Iraq disclosed higher rate of folliculitis (61.9 vs and 28.6%). Jewish patients from North African countries had higher rate of ocular disease and disease severity score was significantly higher in this population. Zouboulis
et al.[
16], analysed the clinical features of 196 BD patients [German (n = 82), Turkish immigrants (n = 86) or immigrants from other countries (n = 28)]. They found a higher rate of ocular disease in South-Eastern European patients (Italy and Greece) compared to South-Western and North European patients. Frequency of ocular lesions was also more frequent in Turkish patients compared to German.
We acknowledge some limitations in the current study. Our analysis was performed as a retrospective review. Socioeconomic differences may exist in our study population, especially between European and African (i.e. North or sub Saharan) patients. However, the time between first symptoms to BD diagnosis was equivalent between groups. Access to medical system does not seem to significantly account for the differences observed because the French social security system covers all medical care. One can hypothesize that genetic factors (i.e. HLA B 51 and/or others) may be involved in ethnic differences with respect to phenotype and outcome of BD. We were not able to collect country of birth of BD patients in a comprehensive manner in order to provide some insight into environment versus genetic influences.
In conclusion, in a French multiethnic country, sub Saharan African BD patients exhibited a worse prognosis. They displayed a higher frequency of CNS involvement and trend toward higher cardiovascular involvement compared to BD patients from Europe or North Africa. Their 15-year mortality rate reached 20% which is 3 times higher than the overall mortality in BD. Taken together, these data suggest that a particular attention should be given to sub Saharan African BD patients.
Competing interests
The authors declare that they have no competing interests.
Authors’ contributions
LS and DS wrote the paper. DS designed the study. MRR performed the statistical analysis. LS, BW, CC, JCP, PC and DS followed the patients. All authors read and approved the final manuscript.