Epidemiological and diagnostic characteristics of systemic lupus erythematosus in a decentralized setting in sub-Saharan Africa

Introduction

Systemic lupus erythematosus (SLE) is a multisystemic autoimmune disease with protean manifestations. It is a complex disease with considerable clinical heterogeneity, morbidity and mortality. It evolves through relapses and remissions. Its heterogeneity often leads to misdiagnosis. It most often affects women, with a sex ratio of 1/9 (1). Clinical manifestations can involve all systems. The etiology remains unknown, and is thought to involve a multitude of genetic, environmental and hormonal factors. Systemic forms with visceral manifestations are the most serious (2). SLE was considered rare in Africa. This is partly due to a lack of awareness regarding its symptoms and diagnosis. Diagnosis is often delayed by the endemic nature of neglected tropical diseases, whose manifestations are frequently confused with those of SLE. There are case series published sporadically (3). Our study will help raise awareness among clinicians about the manifestations of SLE. We conducted a study to assess the epidemioclinical and diagnostic aspects of SLE in Tivaouane.

Material and method

• Study design: We carried out a retrospective study from May 2017 to May 2024 in the medical specialties departments (internal medicine, dermatology) of the Abdou Aziz Sy Dabakh Hospital in Tivaouane. The Department of Internal Medicine offers a wide range of services, including follow-up care, consultations, and hospitalizations, as well as biopsies of the kidneys, liver, skin, and accessory salivary glands.
• Setting and participants: We included all patients who presented with clinicobiological manifestations that led to the diagnosis of SLE in accordance with the ACR/EULAR 2019 and ACR 1997 classification criteria, The skin biopsy revealed lesions characteristic of the disease in some patients. The patients ranged in age from 11 to 70 years. The pediatric population was underrepresented (2)
• Data collection: A data form was drawn up on Kobotoolbox (Kobottolbox is an online data collection software). The data was collected by doctoral students trained on the subject and coordinated by senior university practitioners. All the data collected was checked before final integration. The data consisted of:
• Disease activity was determined using the SLEDAI-2 K score.

No Activity: 0 points; Low/Mild Activity: 1 to 5 points; Moderate Activity: 6 to 12 points; High/Severe Activity: > 12 point

• Analysis: Data exploitation and analysis was made by stata 21.0. Normally distributed continuous variables are reported as mean ± standard deviation (SD). Non-normally distributed continuous variables are expressed as median with interquartile range (IQR). Qualitative (categorical) variables are presented as absolute frequencies and percentages (n, %). We used Fisher's exact test, and statistical significance was observed at a p-value of less than 5%.

Results

Prevalence

Systemic lupus erythematosus (SLE) was diagnosed in 84 patients. This corresponded to a prevalence of 0.15% among 56,000 patients (Followed in the hospital for whole diseases) during the study period. The standard deviation was 0.016%, with a 95% confidence interval (95% CI) of [0.118–0.182%].

Participants' sociodemographic characteristic

The mean age was 36, 67 years old. The female sex was predominated with the sex ratio of 0.09. The sociodemographic data are reported in the Table 1.

Table 1. Socio-demographic data.

Clinical manifestations

Mucocutaneous (54.76%) and articular (64.28%) manifestations were the most frequent. This is illustrated in Table 2.

Table 2. Distribution of clinical signs by body systems and organs.

Clinical manifestations were dominated by general signs with physical asthenia (47.6%), cutaneous signs with non-specific erythematous lesions (15.5%) and erythematosquamous lesions (16.7%). The various specific or related clinical signs encountered in patients treated for SLE are described in Tables 3, 4. Non-specific signs are listed in Table 5. Figures 1,2 illustrate some of the dermatological lesions found in our patients.

Table 3. Significant clinical manifestations of lupus observed in our patients (84 patients).

Table 4. Systemic manifestations of lupus observed in our patients.

Table 5. Other associated and non-specific signs of lupus .

Other signs were found in patients who were suspected of having an autoimmune disease. These non-specific signs are listed in Table 5.

Particularities

Biological and histopathological signs

These included non-specific signs such as anemia (66.1%), an inflammatory syndrome with elevated C reactive protein (CRP) (36.9%), and an accelerated sedimentation rate (ESR) (20.2%). Antinuclear antibodies (ANA) were positive in 71.4% (65 patients) with a mean of 1,007.56 IU ranging from 80 to 10,000. The fluorescence pattern was predominantly speckled (85.9%), followed by homogeneous (5.3%), nucleolar (3.5%), centromeric (3.5%), and cytoplasmic (1.8%) patterns. Other immunological markers, such as anti-CCP (13.1%), rheumatoid factor (10.7%), direct Coombs test (2.4%), and lupus anticoagulant (2.4%), were found to be non-specific for lupus. Skin biopsies were performed on the majority of patients referred from dermatology (21/22)

The biological signs observed in SLE are shown in Table 6.

Table 6. Biological abnormalities observed in patients followed for systemic lupus erythematosus.

Particularities

Additional clinical manifestations were associated with overlapping, systemic or localized autoimmune pathologies, including pernicious anemia, Sjögren's syndrome, and scleroderma. These manifestations included sicca syndrome, cutaneous sclerosis, and melanoderma, accompanied by laboratory findings such as anti-SSA and anti-CCP antibodies

• The two patients with childhood-onset systemic lupus erythematosus (cSLE) were aged 11 and 15 years. Both presented with distinct dermatological manifestations, including malar erythema and specific histopathological features.

• In our cohort, the onset of lupus at an older age was presented with greater disease severity and the presence of cardiopulmonary complications.

Severe manifestations

Severe systemic manifestations were predominantly cardiac (9.5%), renal (7.1%), and neurological (5.9%) complications. The various complications or severe conditions are listed in Table 7.

Table 7. Distribution of patients according to serious symptoms.

Assessment of disease severity

Disease severity was determined using the SLEDAI-2 K score. The mean score was 5.74, ranging from 0 to 19. Disease activity was severe in 22 patients (26.19%), moderate in 19 patients (22.62%), and mild in 43 patients (51.19%)

Associations

The combination of lupus and rheumatoid arthritis was the most frequent, occurring in 6 patients. The various associations are listed in Table 8

Table 8. Distribution of patients according to autoimmune disease associations.

The treatment

The majority of patients (88%) were treated with the immunomodulator hydroxychloroquine, while azathioprine was prescribed for 7.14% of patients. Maintenance therapy was not specified for two patients. The distribution of the various treatment regimens is detailed in Figure 3 and Table 9.

Table 9. Distribution of patients according to other treatments.

Analysis

All cases of cardiac complications occurred in female patients. However, no statistically significant association was found between sex and cardiac involvement (OR=2.59; 95% CI: 0.13–49.5, p = 0.59). The wide confidence interval reflects the small number of events and the imbalance between sexes, which limits the precision of the estimate. Disease severity was more closely linked to delayed diagnosis and was more commonly observed in females (P = 0,18)

Discussion

Limitations: The limitations of this study lie in the fact that it is retrospective in nature. The medical records were not digitized. Not all information was always accurately recorded. The study was conducted in an area with limited financial resources, and patients did not always have the means to undergo sufficient testing or to obtain the necessary medications.

In our study, the prevalence was 0.15%. This is significant given the overall population and the duration of the study. This prevalence was close to that found in the literature (3, 4). It is slightly lower than the prevalence of 0.27% found in the study by Daou et al. across three hospitals in Niamey, Niger (5), and the study by Camara et al., which reported a prevalence of 2.19% over a 5-year period in Abidjan (6). This difference can be explained by our larger study population. Some authors conducted multicenter studies on smaller overall populations, which explains their higher prevalences. For instance, Brah et al. in 2015 reported a prevalence of 0.06%, involving 9 cases of SLE in an internal medicine department over a 10-year period (7).

The mean age in our study was 36.67 years, ranging from 11 to 70 years. This young age of onset has been described in most studies. In a study conducted in Dakar by Fall et al. in 2011, the mean age was 33.9 years (8). Camara et al. found a mean age of 35 years (6). Systemic lupus erythematosus is predominantly a disease of young people, as reported in most epidemiological studies. However, the disease can also affect older adults, sometimes with more severe manifestations (9, 10). It can also occur in the pediatric population (11, 12).

Females were predominantly represented in our study (91.7%). Lupus is a predominantly female condition, as reported in most studies (1, 2). This is also the case for most African series, such as the studies by Daou and Camara in Niamey and Abidjan, respectively (5, 6). However, certain specificities allow for a correlation between the male sex and systemic lupus erythematosus. For instance, in a multicenter study of patients with lupus panniculitis, multivariate analysis revealed that the presence of SLE was significantly associated with the male sex (OR, 8.29; 95% CI, 1.52–45.11; P = 0.01) (13). This female predominance could be explained by certain genetic polymorphisms in women, such as the IKBKE and BANK1 genes observed in Chinese women (14). This strong gender disparity strongly suggests that female sex hormones play a role in the onset and progression of the disease. Indeed, it is widely recognized that estradiol not only enhances the survival of autoreactive B lymphocytes but also stimulates the production of SLE-associated autoantibodies, such as antinuclear and anti-dsDNA antibodies (15).

In our study, joint symptoms (inflammatory polyarthralgia, synovitis) and cutaneous symptoms (erythematous lesions) were among the most frequent, following general signs (asthenia, weight loss, and dyspnea). This symptom distribution is similar to the 2021 study by Camara et al. in Abidjan. In the latter, polyarthritis was the most frequent clinical presentation (85.19%), followed by skin involvement (81.48%) (6). Gassama et al. in Mali found a predominance of cutaneous signs (100% malar rash, 96.97% discoid lupus) and articular signs (75.76%) in a series of 33 patients (16). Another study published by Bija et al. in Cameroon in 2014 revealed a similar distribution. The modes of presentation were dermatological (n = 25), musculoskeletal (n = 23), and renal (n = 7). Clinical manifestations noted during the disease course included fever (n = 20), fatigue (n = 20), weight loss (n = 12), anorexia (n = 11), arthralgia (n = 25), dermatological signs (n = 22), serositis (n = 21), and neuropsychiatric symptoms (n = 8) (17). General signs are a common initial presentation in systemic lupus erythematosus, although they are non-specific. In the study by Eloundou et al. in Cameroon, these included asthenia (73.68%), fever (63.15%), anorexia (52.63%), and weight loss (36.84%) (18). Disease severity was assessed using the SLEDAI-2 K score, indicating severe disease in 26.19% of cases. This severity has been noted in several African studies and may be explained by diagnostic delays due to a lack of awareness of the disease among some practitioners (16–19).

Biological manifestations can be classified into hematological and immunological signs. The most frequent hematological signs in our study were anemia (66.1%), an inflammatory syndrome with elevated CRP (39.6%), an elevated erythrocyte sedimentation rate (ESR) (20.2%), thrombocytosis (5.4%), and leukocytosis (1.8%). The increase in CRP was more strongly associated with the presence of serositis and synovitis. Antinuclear antibodies (ANAs) (71.4%), anti-DNA (8.3%), and anti-ENA antibodies (anti-Sm, anti-U1RNP, anti-Ro52, etc.) were found in most patients. These data are similar to a study by Ndiaye et al. in Dakar on a cohort of 142 patients, which found hematological manifestations in 57% of cases, dominated by anemia (81.7%), leukopenia (19.7%), and thrombocytopenia (14.8%). In their study, ANAs were found in 97.9% of cases, and anti-ENA antibodies in 86.95% (20). Lupus is a polymorphic condition that is challenging to diagnose. We identified several antibodies in our study: ANAs (71.4%), anti-dsDNA (12%), anti-Sm (10.7%), and other anti-ENAs including anti-Ro52 (11.9%), anti-RNP/Sm (4.8%), and anti-U1RNP (3.57%). These various autoantibodies have been similarly reported in several African series (5–7). The presence of these biomarkers, particularly biological ones, facilitates disease classification. These autoantibodies are numerous and, in certain situations, can be used to predict organ damage or assess disease prognosis, with several having been discovered in recent years (21, 22). In certain clinical situations, a skin biopsy can help suggest the diagnosis, as was the case for 25% of the patients with dermatological manifestations.

Severe manifestations of lupus include renal involvement, where proteinuria and renal failure can determine prognosis and survival. Proteinuria was found in sixteen patients in the Eloundou study, and one patient had end-stage renal disease (18). The incidence of lupus nephritis is relatively similar across several countries. The primary lesions of lupus nephritis identified in a study by Mokoli et al. in Kinshasa were focal proliferative glomerulonephritis (GN) (38.5%), diffuse proliferative GN (7.7%), membranoproliferative GN (23.1%), membranous GN (23.1%), and mesangial GN (7.7%) (23). Renal involvement may also present as thrombotic microangiopathy, manifesting as renal failure with hypertension, or as hemolytic uremic syndrome, as reported by Mouna J et al. (24). In our study, some patients had concomitant urinary tract infections that may explain the proteinuria; one patient continued to receive nephrology care after a renal biopsy revealed diffuse glomerulonephritis (Stage IV), for which dialysis was proposed. Cardiac involvement was also found in our study. Lupus can present with heart failure, pericarditis, or rhythm and conduction disorders, which can worsen the prognosis for some patients. Cardiovascular manifestations were found in 46% of cases in a 2016 Dakar study, including pericardial effusion (54%), dilated cardiomyopathy (9%), and pulmonary arterial hypertension (PAH) (22%) (25). It is crucial to recognize these manifestations and optimize management to improve patient prognosis. Neurological manifestations are not uncommon in lupus and can include epileptic seizures, strokes, altered mental status, and peripheral neuropathies, some of which may leave severe sequelae. In our context, these polymorphic clinical presentations can lead to confusion with certain tropical diseases, resulting in diagnostic delays. Some of these clinical presentations (obtundation, psychosis, coma, and motor deficits) were observed during our study. Neuropsychiatric manifestations have been widely reported in patients treated for SLE. Seizures and psychosis were more frequent in early-onset SLE, according to a meta-analysis of 44 studies involving 17,865 lupus patients (26). Pulmonary involvement occurs in 20% to 90% of SLE patients and can carry a poor prognosis, especially in cases of acute respiratory distress. Pulmonary involvement is often correlated with older age and the presence of anti-RNP antibodies (27). It may present as interstitial or alveolar pneumonitis, serofibrinous pleurisy, or atelectasis, and can be complicated by bacterial superinfection. The onset of respiratory distress is a key factor in the severity of this condition and strongly determines the prognosis (28).

Most of our patients were treated with hydroxychloroquine (HCQ), corticosteroids, and conventional immunosuppressants (azathioprine, methotrexate). Maintenance therapy was not specified for two patients. These findings are consistent with the regional literature; HCQ and corticosteroids were the most commonly prescribed drugs in the studies by Brah et al., Eloundou et al., and Gassama et al. (7, 16, 18). HCQ reduces the risk of disease flares and allows for lower corticosteroid doses, while also playing a role in preventing lupus-related thromboembolic disease. It is not contraindicated during pregnancy (28, 29). Depending on the indications, it can be combined with azathioprine or mycophenolate mofetil, which may help reduce disease-related cognitive dysfunction (30). The treatment of lupus has seen significant advances in recent years with the advent of biologics. The development of these molecules is based on a better understanding of the molecular mechanisms involved in lupus pathogenesis. Anti-interferon therapies (anifrolumab, rontalizumab, sifalimumab) have demonstrated good efficacy with a significant corticosteroid-sparing effect, though they carry an infectious risk, such as the development of shingles. Anti-TNF agents (etanercept, etc.) and monoclonal antibodies (belimumab, rituximab, etc.) have also shown strong efficacy; however, patients require close monitoring due to the risk of infection (31–33). Access to these innovative treatments remains limited in Africa, and their high cost restricts their widespread use in tropical regions (18, 27, 34–38).

Conclusion

This study revealed the disparity of lupus disease in its clinical presentation and socio-demographic distribution. However, many diagnostic errors persist due to a lack of knowledge about the disease. It would therefore be useful to provide ongoing training for healthcare personnel and therapeutic education sessions for patients.

Generative AI statement

The author(s) declared that generative AI was not used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher's note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Footnotes

Edited by: Matteo Piga, University of Cagliari, Italy

Reviewed by: Mickael Essouma, Independent Researcher, London, United Kingdom

Medhat Taha, Mansoura University, Egypt