Behçet disease (BD) is classified as an inflammatory vascular condition, affecting all kind and size of vessels.1 Although BD is distributed worldwide, its demographic and clinical features are strongly associated with the ethnic group and the geographic area involved.2

The highest prevalence of BD was observed in Asia, especially in Turkey, Israel, Iran, and Japan (prevalence range: 13–420/100,000 inhabitants). In Southern and Western European countries (Italy, Spain, and France) the prevalence rate was intermediate (prevalence range: 1.5–16/100,000 inhabitants), while Northern European countries (Sweden, England, and Scotland) and the United States reported the lowest prevalence (prevalence range: 0.3–5.2/100,000 inhabitants).3

BD usually begins in the third or fourth decade of life, with onset in childhood or late adulthood being rare. The female/male ratio varies among different ethnic groups and geographic regions, with male predominance in Middle Eastern countries and female predominance in European and North American countries. Most series agree that male gender has a more severe disease, regardless of demographic characteristics.2

The disease has a broad spectrum of clinical features and is characterized by periods of exacerbation and remission. Its phenotypic expression also varies according to the ethnic group and country. Skin and mucosal lesions comprising oral and genital ulcers, papulopustular lesions, and erythema nodosum-like lesions are observed in most patients. On the other hand, uveitis, venous thrombosis, arterial thrombosis and aneurysms, nervous system and gastrointestinal involvement occur in a smaller number of patients.4

The genetic predisposition for BD is suggested by the higher concordance for BD in monozygotic twins compared to dizygotic twins.5 The genetic approach identified associations between different genetic markers and BD, with HLA-B*51 allele being the strongest susceptibility factor for the disease. It is estimated that carriers of HLA-B*5/B*51 have a pooled odds ratio of 5.78 for the development of BD (95% confidence interval: 5–6.7) compared with non-carriers. Moreover, the population-attributable risk for BD concerning the presence of HLA-B*5/B*51 within different geographic areas was estimated at 32–52%.6

Limited data exist on the demographic, clinical features and the HLA-B*51 frequency of BD in Latin American countries. This region is characterized by being multi-ethnic (White, Mestizo, African Latin Americans and Pure Amerindians) and multicultural. Data from the GLADEL (Grupo Latino Americano De Estudio de Lupus) cohort demonstrated that patients with mestizo systemic lupus erythematosus have significant differences (demographic and clinical characteristics) compared to white patients. Indeed, these differences are associated with a worsening overall prognosis for mestizo SLE patients.7 It is therefore necessary to better describe the demographic and clinical characteristics of BD patients in Latin America countries that may be unique to this population.

The aim of this work was to describe through a systematic literature review (SLR) the demographic and clinical features, as well as the frequency of HLA-B*51 allele in BD patients in Latin America countries.

An SLR was carried out according to PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analysis) guidelines without performing a meta-analysis.8

The literature search yielded 86 articles, as shown in Fig. 1. A total of 74 articles were excluded for several reasons, and 12 were considered for analysis in this SLR.3,10–20 Seven articles were from Brazil, two from Colombia and one from each of the following countries: Argentina, Chile, and Mexico. Ten articles refer to the demographic and clinical findings of BD patients (Table 1). Two articles deal with the prevalence of BD,19,20 and three articles with the frequency of HLA-B*51 in BD patients10,17,18 and are discussed separately in the corresponding section.

Fig. 1.

Chart flow: identification of studies.

(0,15MB).

Latin America is an ethnic-geographical concept that emerged in the 19th century to identify a region of the American continent where most languages are derived from Latin (mainly Spanish or Portuguese and, to a lesser extent, French). The concept was introduced to differentiate between regions of the continent with Latin, Catholic and mestizo influence, and those characterized by the Anglo-Saxon, Protestant culture, with little or no miscegenation. However, Latin America is a vast region with heterogeneous geographic, ethnic, socioeconomic, and cultural characteristics.21 This fact could modify features of BD, such as its prevalence, genetics, clinical findings, and prognosis.

Through a systematic literature review, we described the demographic, clinical features, and HLA-B*51 allele frequency of BD in Latin American countries.

Data on the prevalence of BD in Latin American countries is scarce, with only 2 studies reported on this topic.19,20 In 2008 Tunes et al. from Brazil reported a prevalence of 0.3/100,000 inhabitants in Bahia State (Northeast Brazil), with a total population of 385,109. The study objective was to evaluate the prevalence of BS in patients with recurrent aphthous ulceration (RAU) followed up in a specialized ambulatory clinic from a university institution. Furthermore, cases of BS with less symptomatic oral aphthae, and hence not meeting the RAU criteria are not included in this study. Therefore, the authors stated that the prevalence of BD in the population could be underestimated due to the “excluded third party” bias.19 More recently, Fernández Avila et al. from Colombia reported a prevalence of 1.1/100,000 inhabitants. Their study was designed to establish the prevalence of BD in Colombia based on the administrative clinical records from the Ministry of Health.20 These studies showed a low prevalence of BD in our region, similar to that in North America (0.3–5.2/100,000) or Northern European countries (prevalence range 02–4.9/100,000).3

BD demographic and clinical features differ worldwide.2 In this SLR, the mean age at disease onset was 33 years (range 26.3–42.1), which is in agreement with previous reports of disease onset in the third/fourth decade.22 The female-to-male ratio was 1.4:1; this slight female gender predominance is consistent with the results described in North America and Western European countries.23 The frequency of major clinical manifestations suggests that BD pattern in Latin America is similar to that described in other regions where this condition is endemic. In order of frequency, oral and genital ulcers, skin, eye, and joint involvement were the most commonly observed manifestations. Similar results were shown in large series of BD patients from endemic areas. Gastrointestinal (GI) involvement is an uncommon manifestation in BD, except in Japan, where the reported prevalence is higher than in the rest of the countries/regions worldwide.24 Interestingly, in this SRL, the frequency of GI manifestation (17.5%) was even higher than that reported in Japan (14%). Soto-Vega et al. reported that 7 of 32 (22%) BD patients had GI lesions; however, the study failed to outline the type of lesions they presented. Toro Giraldo et al. reported a high frequency of GI involvement (8/20; 40%). The type and frequency of manifestation were abdominal pain (25%), mucosal ulcerative lesion (25%), lower GI bleeding (20%), recurrent diarrhea (15%), and intestinal perforation (5%).3 Tunes et al. reported a very high frequency of GI manifestation (10/16; 65.2%); however, except for gastroduodenitis (18.8%), no other manifestation was mentioned in their cohort.13

HLA-B*51 allele is known to be an important susceptibility gene in BD, regardless of geographical region and ethnicity, with a close correlation between the geographical distribution of HLA-B*51 and prevalence of BD. De Menthon et al. in a systematic literature review and meta-analysis observed that the pooled prevalence of HLA-B*51/B*5 allele was 57.2% (95% CI 53.4–60.9) and 18.1 (95% CI 16.1–20.3) for BD patients and healthy controls, respectively. The OR of HLA-B*51/B*5 allele carriers to develop BD compared with non-carriers was 5.78 (95% CI 5.00–6.67). The subgroup analyses stratifying studies by geographic locations showed the highest prevalence of HLA-B*51/B*5 allele in BD patients from Eastern Asia, Middle East/North Africa, and Southern Europe (prevalence range 55–63.5%).6 The OR range (5.18–7.20) for these regions was consistent with the overall OR. Although the prevalence of HLA-B*51/B*5 allele in BD patients (39%) was lower in the Northern/Eastern Europe region, the OR (5.31; 95% CI 3.35–8.40) continues to be consistent with the overall OR. On the other hand, in the North America region, the prevalence of HLA-B*51/B*5 allele (34.2%) was even lower than that in Northern/Eastern Europe and the OR (2.35; 95% CI 0.56–9.82) showed no association with BD presence. The Latin American region was underestimated in this meta-analysis due to the lack of scientific studies evaluating the prevalence of HLA-B*51 in BD patients. In fact, only one Latin American study carried out in Mexico had been included. Lavalle et al. in a small study (10 BD patients and 105 healthy controls) found that HLA-B*5 (HLA-B*51 is a split product of HLA-B*5) allele was more frequent in Mexican Mestizo patients with BD than in ethnically matched controls (70% vs. 31%, p<0.05).25 More recently, 3 Latin American studies reported the prevalence of HLA-B*51 allele in patients with BD and in healthy controls. Soto Vega et al. found no difference in the prevalence of HLA-B*51 allele between Mexican Mestizo BD patients and healthy controls (9% vs. 5%; OR 1.88, 95% CI 0.58–5.93). On the other hand, studies from Brazil and Argentina revealed a prevalence of HLA-B*51 in BD patients of 30.1% and 38%, respectively; these results are similar to those reported in the Northern/Eastern Europe region.3,10,17,18 Both Brazilian and Argentine studies demonstrated an association between HLA-B*51 allele and BD presence (OR=2.34, 95% CI 1.32–4.14 and OR=3.75, 95% CI 1.71–8.19, respectively).17,18 It is noteworthy that the difference between Soto Vega study and Brazil and Argentina studies could be ethnic. While the Mexican study included all mestizo patients, Brazilian and Argentinean studies were comprised of mestizos and other ethnicities (mainly Caucasian and Amerindian).10,17,18

The prevalence of BD in the Latin American region, as well as the prevalence of HLA-B*51 allele, seems to be low; however, the strength of association between HLA-B*51 and BD remains high in our population. The key clinical features of BD are similar to those in countries/regions where BD is endemic. Finally, more studies of the prevalence, demographic and clinical features of BD in Latin American countries are required to confirm these findings.

The authors declare they have no conflict of interest.