We conducted a retrospective, descriptive cohort study over a 5-year period from August 29, 2016, to August 18, 2021, within the Internal Medicine Department of the Dalal Jamm National Hospital Center (CHN Dalal Jamm) in Dakar, Senegal. The end of this period coincided with the outbreak of SARS-CoV 2 pandemic, during which the department was partially integrated into the epidemic treatment center, disrupting the usual rhythm of outpatient visits and hospitalizations.

A total of 1,136 medical records of patients seen in the outpatient clinic or hospitalized for a systemic condition were initially screened. Among these, 130 records were excluded because they were incomplete or did not allow for the extraction of basic demographic or diagnostic data. This selection process resulted in a final cohort of 1,006 unique patients, ensuring the transparency of our sampling frame and final cohort selection. Inclusion was based on diagnostic confirmation according to validated international classification criteria [4]:

1) Rheumatoid arthritis (RA): 2010 ACR/EULAR criteria [5][6].

2) Systemic lupus erythematosus (SLE): 2019 ACR/EULAR criteria or SLICC criteria [7]-[9].

3) Primary Sjögren’s syndrome (PSS): 2016 ACR/EULAR criteria [10].

4) Systemic scleroderma (SSc): 2013 ACR/EULAR criteria [11].

5) Idiopathic inflammatory myopathies (IIM): Hoogendijk or Troyanov criteria [12][13].

6) Familial mediterranean fever (FMF): Tel Hashomer diagnostic criteria [14].

7) Ankylosing spondylitis (AS): Modified New York criteria [15].

8) Juvenile idiopathic arthritis (JIA): Edmonton criteria [16].

9) Systemic vasculitides: Definitions from the 2012 Chapel Hill Consensus Conference [17].

10) Sarcoidosis: Consensus clinical, radiological, and histopathological criteria [18].

Data were collected using a standardized questionnaire assessing demographic information, reasons for consultation, time to diagnosis, comorbidities (particularly cardiovascular risk factors), gynecological and obstetric history, and treatment regimens received.

The time to diagnosis was defined as the interval between the onset of the first clinical symptom and formal diagnostic classification based on international criteria. Symptom onset was retrospectively identified and extracted from the history of the disease section of the medical records, as documented during the patient’s initial internal medicine consultation. To address uncertain or partially missing dates, a standardized imputation rule was applied: when only the year of onset was known, the onset was set to the midpoint of that year (July 1st); if only the season or month was recalled, the midpoint of that specific period was utilized.

Statistical analysis was performed using Excel and SPSS software. Quantitative variables are expressed as means and standard deviations, qualitative variables as counts and percentages. Bibliographic management was conducted using Zotero software according to the Vancouver system. The study received the required administrative approvals, and patient anonymity was strictly maintained.

A total of 1,006 patients were included. The mean age of the cohort was 48.11 years, with a high concentration of patients (74.64%) in the 25 - 64 age group. There was a clear predominance of females (82.03%), resulting in a sex ratio (M/F) of 0.219. The time to diagnosis was documented for 675 patients. The distribution was heavily skewed with a mean of 4.34 ± 5.88 years, a mode of 1 year, and a median of 3 years (range: 0 to 46 years). Subgroup variations are presented in Figure 1 and Table 1 (p. 3). The mean time to diagnosis (from the first symptom to formal classification) was 4.34 years, varying across pathological subgroups (Figure 1) (Table 1):

Figure 1. Variation in diagnostic delays according to the spectrum of systemic diseases.

Table 1. Epidemiological aspects according to the spectrum of systemic diseases.

Note: *Legend: APS/SAPL: Antiphospholipid Syndrome; SD: Standard Deviation; IBD: Inflammatory Bowel Disease.

The reasons for consultation were dominated by joint manifestations (83.73%), primarily of the polyarthritis type (60.6%). Cardiovascular risk profiles were comprehensively documented for a subset of 594 patients; data were missing for the remaining 412 patients primarily due to the acute/emergency context of certain admissions or incomplete long-term retrospective charting. Within this analyzed subset of 594 patients with a complete cardiovascular profile, 37.21% had at least one cardiovascular risk factor: hypertension (25.79%), diabetes (7.57%), dyslipidemia (2.35%), stroke (0.33%), and peripheral arterial disease (0.33%). In terms of gynecological and obstetric history, 6.38% of the women in the series (n = 51) reported at least one history of spontaneous abortion [19][20].

The cohort was divided into defined immunopathological subgroups (Figure 2) (Table 2):

1) Systemic autoimmune diseases (67.1%, n = 675): Largely dominated by rheumatoid arthritis (49.7%, n = 499), followed by overlap syndromes (6.5%, n = 65), systemic lupus erythematosus (3.4%, n = 34), primary Sjögren’s syndrome (3.2%, n = 32), inflammatory myopathies (2.7%, n = 27), systemic sclerosis (0.6%, n = 8), primary SAPL (0.4%, n = 4), multiple autoimmune syndromes (0.3%, n = 3), Sharp syndrome (0.2%, n = 2), and atrophic polychondritis (0.1%, n = 1).

2) Undifferentiated connective tissue diseases: These accounted for 10.9% (n = 110) of the entire cohort.

3) Autoinflammatory diseases (19.4%, n = 195): Primarily composed of polygenic forms: ankylosing spondylitis (8.8%, n = 89), gout (6.0%, n = 60), rhizomelic pseudo-polyarthritis (2.5%, n = 25), juvenile idiopathic arthritis (1.5%, n = 15), systemic IBD (0.3%, n = 3), adult-onset Still’s disease (0.1%, n = 1), and psoriatic arthritis (0.1%, n = 1). A single case of a monogenic form was identified: Familial Mediterranean Fever (FMF) (0.1%).

4) Systemic vasculitides (2.1%, n = 21): Including Behçet’s disease (1.4%, n = 14), granulomatosis with polyangiitis (n = 2), microscopic polyangiitis (n = 2), giant cell arteritis (n = 1), ANCA-associated vasculitides (n = 1), and nodular periarteritis (n = 1).

5) Granulomatoses (0.5%, n = 5): Represented by Kimura disease (n = 2), sarcoidosis (n = 2), and Rosai-Dorfman disease (n = 1).

Figure 2. Distribution of patients according to the spectrum of systemic diseases.

Table 2. Distribution of patients by diagnosis.

Long-term maintenance treatment was initiated in 583 patients. The remaining 423 patients did not receive long-term csDMARDs or maintenance therapy because they either defaulted immediately after diagnostic confirmation, elected to pursue traditional alternative medicine, or presented with self-limiting acute conditions requiring only short-term symptomatic interventions.

Among the 583 treated patients, Methotrexate was the most commonly prescribed drug (54.20%, n = 316), followed by hydroxychloroquine (43.34%, n = 253), sulfasalazine (12.03%, n = 70), allopurinol (7.03%, n = 41), azathioprine (4.46%, n = 26), mycophenolate mofetil (0.51%, n = 3), and leflunomide (0.34%, n = 2). Dual therapy was utilized in 22.47% of these cases, whereas triple therapy was observed in only two patients. Access to biologic therapy (Rituximab) remained rare (0.17%, n = 1) [21] (Table 3).

Table 3. Distribution of patients by monotherapy maintenance treatment.

*Note:Percentages sum to greater than 100% due to patients undergoing overlapping dual/triple regimens.

Our study describes one of the largest single-center cohorts of systemic diseases in West Africa. Our results demonstrate that the immunological landscape in tropical settings is far broader and more complex than previously recognized. While 20th-century paradigms suggested a relative rarity of autoimmune diseases among African populations with dark skin, the analysis of these 1,006 cases documented over 5 years reflects an increasing detection rate in specialized clinical structures [22].

Consistent with global data, RA is the most common condition, accounting for nearly half of our cohort (49.7%). Its prevalence aligns with Kane’s multicenter studies in Dakar [22]. It is interesting to note geographical and recruitment variations on a continental scale: while our series places lupus (3.4%) and undifferentiated connective tissue diseases (10.9%) behind RA, series in Ouagadougou report a predominance of systemic sclerosis [23], and teams in Abidjan describe lupus as the leading condition [24]. These discrepancies likely reflect the density of specialists in rheumatology or internal medicine across different capitals rather than actual genetic gradients [25].

The mean age (48.11 years) and the strong female predominance (sex ratio of 0.219) align with data from the international literature, confirming the role of sex hormones in modulating the immune response. Furthermore, the high prevalence of cardiovascular risk factors (37.21%), particularly hypertension (25.79%), underscores that the development of accelerated atherosclerosis induced by chronic systemic inflammation represents a major prognostic challenge in Africa [6][20].

The overall median diagnostic delay of 3 years (mean: 4.34 years) constitutes a major obstacle to optimal management, as uncontrolled chronic inflammation leads to irreversible structural or visceral damage [22]. However, detailed analysis by disease reveals that this diagnostic delay is particularly extreme for rare or underrecognized conditions in our region. In our series, the longest diagnostic delay was observed for a monogenic autoinflammatory condition, Familial Mediterranean Fever (FMF), where the gap between the age of symptom onset in the patient and the age at diagnosis reached 20 years [26].

This delay is considerably longer than those reported in countries where the disease is classically described. In Morocco, a country with a high prevalence of this condition, the average diagnostic delay within a national cohort is only 10.75 years [27]. These results clearly demonstrate that diagnostic delay for FMF remains the norm in our context [26]. This critical delay is largely due to a lack of awareness of the disease among our frontline healthcare professionals, who rarely consider monogenic autoinflammatory syndromes when faced with recurrent febrile episodes or serositis in tropical settings, mistakenly preferring to perform multiple tests for infectious or parasitic causes.

While FMF illustrates the complexity of monogenic forms linked to innate immunity [4], the autoinflammatory diseases (AIDs) in our cohort are largely dominated by polygenic entities (19.4% of the overall cohort) [22]. Ankylosing spondylitis (8.8%) and gout (6.0%) constitute the main core of these conditions [22]. Long underestimated among Black Africans, ankylosing spondylitis emerges here as a major reason for seeking care in tropical rheumatology, despite the historically low prevalence of the HLA-B27 allele in this subregion, suggesting other polygenic pathways or environmental factors specific to the tropical setting [28].

Gout, for its part, represents a polygenic metabolic IAM whose rising incidence reflects the Westernization of diets and the increase in metabolic syndromes in urban Africa [29]. Finally, polymyalgia rheumatica (PPR) (2.5%) demonstrates that inflammatory conditions in the elderly are actively emerging in Senegal, in parallel with increasing life expectancy. These polygenic forms share a critical challenge in tropical regions: their tendency to mimic septic arthritis or post-infectious reactive arthropathies, frequently leading to inappropriate antibiotic prescriptions and delaying the initiation of csDMARDs or colchicine [9].

Although granulomatous diseases account for only 0.5% (n = 5) of our cohort, their relevance in tropical internal medicine is critical due to the diagnostic pitfalls they share with granulomatous infectious diseases, foremost among which is tuberculosis. Sarcoidosis (2 cases) and Rosai-Dorfman disease (1 case) are characterized by lymphadenopathy and systemic involvement, the clinical and radiological presentation of which closely mimics lymph node or multifocal tuberculosis. In West Africa, identifying the precise origin of such respiratory signs remains a cornerstone of clinical practice, as pleuropulmonary manifestations are common features of underlying connective tissue diseases [30]. In tropical settings, the practice of routinely treating all granulomatous diseases as tuberculosis (due to endemicity) exposes these patients to unnecessary drug toxicity and delays the initiation of life-saving corticosteroid therapy [18].

The presence of two cases of Kimura’s disease, a cutaneous-lymph node eosinophilic granulomatosis that is extremely rare in dark-skinned individuals, illustrates the extreme diversity of our center’s patient population [19]. When faced with a presentation of skin nodules and lymphadenopathy associated with massive hypereosinophilia in a tropical setting, the clinician legitimately and primarily considers filariasis or another invasive helminthiasis. The absence of microfilariae and histopathological confirmation of a granulomatous infiltrate rich in immunoblasts and eosinophils without caseous necrosis allow the diagnosis to be revised in favor of Kimura’s disease, thereby avoiding repeated courses of ineffective anthelmintics [19].

Within our vasculitis subgroup (2.1%, n = 21), Behçet’s disease was the predominant form with 14 cases (66.6% of vasculitides), a finding fully consistent with Kane’s observations [22]. To contextualize these data, it is worth referring to the seminal work by Ndongo et al. on systemic vasculitides in Senegal [31]. In their study of 27 cases collected in hospitals in Dakar, the authors had already highlighted the overall rarity of primary systemic vasculitides in sub-Saharan Africa, while confirming the predominant role of Behçet’s disease within this spectrum [31].

A crucial element shared by the observations of Ndongo et al. concerns the importance of secondary vasculitides in our setting [32]. In tropical African settings, presentations of vasculitis occur very frequently during the course of major, long-standing underlying connective tissue diseases [32]. This is the case with rheumatoid vasculitis complicating severe and neglected rheumatoid arthritis or lupus vasculitis, which is particularly favored by delayed diagnosis [32]. Furthermore, clinicians must distinguish these forms of vasculitis from those associated with chronic endemic infections (e.g., hepatitis B and C viruses, HIV), which often present as cryoglobulinemia or secondary leukocytoclastic vasculitis mimicking primary forms [32]. This phenotypic duality requires absolute diagnostic rigor to avoid serious therapeutic errors based on indiscriminate immunosuppression.

The main strength of this study lies in its large sample size (1,006 patients) recruited from a single university center, ensuring relative homogeneity of clinical criteria. Conversely, its retrospective nature results in missing data. Furthermore, technical limitations (unavailability or high cost of certain specific immunological tests, such as detailed ANCA testing or genetic testing for autoinflammatory diseases) may have limited diagnostic accuracy, explaining the rate of connective tissue diseases classified as “undifferentiated” (10.9%).

Additionally, because our department serves as a major national tertiary referral center, the study cohort is subject to a significant selection bias. The observed cases represent hospital-based frequencies within a specialized setting rather than the true prevalence or incidence of these conditions in the general population.