To our knowledge, this study is the first in sub-Saharan Africa to specifically describe the profile of HLA-B27-positive ERA. Previous studies have described ERA in the context of juvenile idiopathic arthritis using ILAR criteria without HLA-B27 screening [14, 15].

In Senegal, three ethnic groups predominate Wolof (43%), Fula (Peul) (24%) and Serer (15%) representing a total of 82% of the Senegalese population. Our study shows that HLA-B27-positive ERA mainly affects the Fula ethnic group (87%), even though this ethnic group represents only 24% of the population. This result was linked to the high prevalence of HLA-B27 (6-7.8%) in this ethnic group compared with the rest of the sub-Saharan African population, which is generally less than 1% [6,7,8]. Another factor is the high frequency of endogamy (consanguineous marriage) within this ethnic group, as illustrated in Fig. 1.

In this study, there was a male predominance (64.5%) which was generally consistent with the findings of other HLA-B27-positive ERA cohort studies (Table 3). The mean age of onset in our patients was approximately 12 years, which was similar to that reported in the literature [5, 16,17,18,19]. However, an earlier median age has been reported in a Turkish study (10 years) [20].

The long duration of symptoms (median: 6.0 years) is linked to several factors which constitute barriers to early diagnosis and treatment in our context. We can suggest three possible explanations for that. Firstly, some patients initially favour the use of traditional medicines (phytotherapy), before consulting in the hospital at a later stage. Secondly, the unawareness of this disease in primary health care facilities, which treat it symptomatically (analgesics) without early referral to rheumatology department. Thirdly, the scarcity of rheumatology specialists in Senegal. This delay highlights the need to raise awareness of this diagnosis and facilitate earlier recognition and referral to rheumatologist.

The median age of our patients (19 years) at diagnosis reflects the progressive nature of the disease, starting in childhood and continuing into adulthood. This confirms that ERA represent a continuum with adult spondyloarthritis [4].

Remarkably, approximately 22% of patients reported a family history of SpA, supporting a major role for genetic factors. This result was in agreement with the Indian study [5].

In our cohort, peripheral arthritis and enthesitis were the most common presenting features at disease onset. This finding is related to the natural course of the disease. Spondyloarthritis usually begins in children with peripheral arthritis predominantly involving the lower limbs and/or with peripheral enthesitis rather than axial arthritis [1, 21, 22].

Enthesitis is a hallmark feature of ERA [23, 24] and reported at the time of diagnosis in 44.4–70% of patients with HLA-B27 positive ERA [5, 20, 25]. In the current study, we found a relatively high rate (83.8%) of enthesitis compared to previous studies.

The majority of our patients with peripheral arthritis had predominantly knee and ankle joint involvement. These findings were in accordance with previously published HLA-B27 positive ERA series [17, 19, 20]. The frequency of sacroiliitis at disease presentation in our study (88%) was higher than in reported studies (51.3 to 83.3%). This frequency is linked to the long duration of the disease (6 years) in our patients in contrast with previous studies.

Uveitis was the most common extra-articular manifestation of HLA-B27-positive ERA [16, 20]. In this current study, the frequency of uveitis was (22.5%). This result was in agreement with the Turkish study (22.2%). However, lower prevalence was reported in Indian (14.7%) and American (13%) studies [5, 26].

Concerning the assessment of enthesitis/spondylitis related arthritis, we opted to use the BASDAI, ASDAS and BASFI to assess patients with ERA. Given that the majority of patients presented the advanced form of ERA known as ankylosing spondylitis or radiographic axial SpA. Indeed, Batthish et al. demonstrated that both the BASDAI and BASFI showed excellent intra-rater reliability in a cohort of ERA patients [27]. Our study demonstrates that there is a positive correlation between disease activity (ASDAS, BASDAI) and BASFI. This indicates that active inflammation has an impact on physical function.

Regarding drug therapy, although biological agents (TNFα inhibitors) are included in the WHO (World Health Organisation) essential medicines list for paediatric rheumatology [28], they are unavailable in our context. The therapeutic options currently available are NSAIDs, glucocorticoids and conventional synthetic disease-modifying anti-rheumatic drugs (methotrexate, sulfasalazine). In Senegal, we use a combination of methotrexate/sulfasalazine and NSAIDs associated with glucocorticoids (intra-articular injections).

Several limitations of our study should be noted. This cohort was retrospective and other extra-articular manifestations such as cardio-vascular involvement was not studied. Magnetic resonance imaging was not used in our study due to a lack of resources. The disease activity at diagnosis was probably under-evaluated because some patients initially received analgesic treatment in primary health before referral.

Despite the study’s limitations, this research is highly relevant to future studies on juvenile spondyloarthritis in sub-Saharan African populations, which are generally under-represented in medical research.