In our cohort of patients, demographic data were comparable with other published literature describing cSLE cohorts in Asia.11–13 Our large cohort exhibited similar female to male ratios, along with the typical female preponderance owing to the role of hormonal factors in lupus disease expression. Females represented 89.1% of the cases with a female to male ratio of 8.2:1, which is well within the common incidence reported for cSLE in asian populations.11–13 The female to male ratio in cSLE is generally found to be significantly less than in aSLE. A higher proportion of female patients is often reported in aSLE, with 96% female patients (female to male ratio of 23.9:1) reported in an adult Filipino SLE population.14 Age-related gender differences normally seen in other pediatric cohorts were similarly seen in our cohort, with increasing ratios observed in those with pubertal or post-pubertal onset.12,15−16
Comparable to other studies as well is the mean age at diagnosis at 13 years old. We reported 3 yrs old as the youngest age at initial diagnosis. This again is similar to other series but relatively younger compared to a previously published study in the Philippines which reported the youngest age at 8 yrs old.11
The frequencies of systemic involvement at diagnosis are summarized in Table 3. In our patients, the most common manifestations at the time of diagnosis were fever, anemia, and arthralgia. Fever as a common presenting manifestation is comparable to data described by local, as well as Malaysian and Portuguese cohorts.11–12, 17 However, contrary to the Filipino cSLE cohorts described in previously reported studies, malar rash and photosensitivity were not as common in our patients.11,18
Table 3
Comparison of systemic involvement at diagnosis with other cSLE cohorts.
| Present Study | Gulay and Dans, 2011 | Cabral, 2013 | Aggarwal, 2018 | Lim, 2020 |
No. of Patients | 304 (%) | 78 (%) | 56 (%) | 273 (%) | 141 (%) |
Country | Philippines | Philippines | Portugal | India | Malaysia |
Mucocutaneous | 55.9 | 91.0 | - | - | - |
Malar rash | 35.5 | 65.3 | 10.7 | 23.5 | 56.0 |
Discoid rash | 6.3 | 32.0 | 1.8 | - | 15.6 |
Photosensitivity | 3.3 | 55.1 | 1.8 | 42.8 | - |
Oral ulcers | 24.0 | 53.8 | - | 23.0 | 48.9 |
Alopecia | 17.8 | 39.7 | - | 60.8 | 41.8 |
Hematologic | 56.9 | 51.2 | 14.3 | - | - |
Anemia | 45.0 | 8.9 | 1.8 | 36.1 | 22.0 |
Leukopenia | 20.0 | 15.3 | - | 24.1 | 51.1 |
Thrombocytopenia | 19.4 | 14.1 | 12.5 | 21.2 | 41.8 |
Musculoskeletal | 43.4 | 53.8 | - | - | - |
Arthritis | 27.3 | 21.7 | 41.1 | 69.2 | - |
Renal | 42.1 | 62.8 | 1.8 | 48.7 | 39.7 |
Neuropsychiatric | 3.6 | 30.7 | 1.8 | 22.7 | 16.3 |
Serositis | 8.6 | 26.9 | 1.8 | 4.2 | 18.4 |
ANA are present in more than 99% of children with SLE19 and so in its absence, the diagnosis of SLE becomes questionable. However, as in most series, our cohort also includes a patient with negative ANA. This was a 17-year-old female who presented with alopecia, edema, and low C3. The diagnosis of SLE was confirmed with a positive anti-dsDNA. Some of the patients in this cohort no longer had ANA tests done. These are patients who initially presented with clinical manifestations that made the probability of SLE very high, hence were directly requested anti-dsDNA to confirm the diagnosis.
In general, the prevalence of antiphospholipid antibodies in SLE is about 20-60%.20 Other cSLE cohorts studied by Lim et al. and Cabral et al. report of lower incidences at 14.9% and 16.4% respectively.12,17 In our cohort, its incidence is much lower at 2.3%, but this is primarily because antiphospholipid antibodies are not routinely checked unless the patients present with symptoms.
Disease activity as measured with SLEDAI-2k scores was generally high at diagnosis in our cohort of patients. This corresponds to data showing that cSLE is more aggressive and severe compared to aSLE.21 The high proportion of patients in our cohort presenting with significant major organ involvement at diagnosis is likewise seen in other Asian studies, though worse than reported in Caucasian cohorts.22–23
One of the most important predictors of poor outcomes in lupus is renal involvement. Manifestations can vary from asymptomatic proteinuria to end-stage renal disease (ESRD). Generally consistent with other reported literature, the predominant kidney biopsy result of lupus nephritis patients in this cohort were class III and IV. Although most commonly reported, this may not be representative of the renal disease in our cohort, considering that not all patients with renal involvement have kidney biopsies done, usually due to financial constraints. For lupus nephritis, particularly class III and IV, aggressive treatment is recommended to prevent progression of disease. Overall, only 1% ended up with ESRD. This low incidence may be reflective of the prompt and adequate management of active renal disease in our center.
Several studies have looked at the pattern of accumulated damage amongst their cohorts. Across most published studies, the most frequently involved organ systems are the ocular, neurological, renal and musculoskeletal systems. Similar to the study by Tucker et al.,24 our cohort had the highest percentage of damage in the renal domain (20.7%), followed by the neuropsychiatric (8.6%) and the musculoskeletal (4.6%) systems. This is in contrast to other series where rates of renal damage were lower.25 Based on the Systemic Lupus International Collaborating Clinics/American College of Rheumatology (SLICC/ACR) Damage Index, renal organ damage is defined as either proteinuria ≥3g/24 hrs, estimated GFR <50%, or end-stage renal disease. In our cohort, among the 67.8% of patients with renal involvement, 25.7% developed nephrotic-range proteinuria (≥3g/24 hrs). Only 3.4% had an eGFR of <50%, and only 1.5% had end-stage renal disease. This cohort of patients must be followed-up into adulthood considering the significance of renal damage amongst adult Asian lupus patients.25–26
In contrast to other series, our cohort had a markedly lower rate of ocular damage (3.0%) despite having patients exposed to moderately high doses and prolonged duration of corticosteroids due to high disease activity, and universal use of hydroxychloroquine.
Therapy given to children in our study is similar to that carried out in patients from aSLE cohorts. This is reflective of how pediatric SLE management is largely based on the clinical experience and large randomized controlled trials in adults. In our cohort, more patients received IV CYC than MMF as induction therapy. The choice of induction treatment was based upon the individual physician’s choice. Previous cSLE studies show IV CYC and MMF to be comparably efficacious with regards to treatment response, damage accrual and time to next flare.27 A prospective comparison of IV CYC versus MMF induction treatment is highly needed to better formulate future lupus nephritis treatment protocols for children.
In aSLE patients with severe or refractory disease, RTX has been used as an adjunctive therapy with good results. However, few studies have been conducted in children and none of these were randomized controlled trials.28 RTX was given to 19 patients in our cohort, mostly for lupus nephritis but also for refractory thrombocytopenia. RTX has been shown to be safe and effective for treating the renal and hematologic manifestations of cSLE, especially in terms of disease activity, immunologic measures, and steroid-sparing effect. Unfortunately, there was failure to achieve remission in most of our lupus nephritis patients who were given RTX. This could be due to the fact that all of them started off with really bad nephritis, and all failed with CYC. Treatment failure of RTX in lupus nephritis is reported in up to 43% of patients in some studies.28 In addition, even if remission is attained with immunosuppressive drugs, nephritis relapse can occur in 35% of responders and seems to be a determining factor in end-stage kidney disease in children.29
Several medications used in the treatment of cSLE may produce severe complications including corticosteroid-related adverse effects and infections from cyclophosphamide. Considering how the medications given may play a factor in the development of infections in SLE, the balance between benefits and side effects should always be considered in selecting options for cSLE treatment. For both children and adults with SLE, infection remains the most common cause of morbidity and mortality,30 as was observed in our patients. Severe infection, along with active disease, were the causes of mortality in our cohort. Most of these patients belonged to low income families, where poor follow-up and treatment compliance are major contributing factors.
Study limitations include its retrospective nature. Data collected and analyzed were purely sourced from information gathered in the chart review, and is highly reliant on the accuracy of documentation in patients’ records. Although our cohort is one of the largest ever studied for cSLE patients in the country and in Southeast Asia, as a single-centre study, ethnic composition of the cohort may be homogenous and therefore general applicability of data may be limited. Primarily due to financial constraints, another limitation of the study was the inability to have complete antibody profiles tested for all patients. The same way kidney biopsies were not done in all patients with renal involvement. This may account for variations in the reported frequencies among different studies.