Impact of high disease activity on damage accrual and disease outcomes in childhood-onset systemic lupus erythematosus Authors: Ninlapat Jidmahawong, Butsabong Lerkvaleekul, Kwanchai Pirojsakul, Soamarat Vilaiyuk Published: 2026 ABSTRACT ObjectiveThe impact of the duration and cumulative high disease activity on damage accrual in childhood-onset SLE (cSLE) is less well established. This study aimed to assess the cumulative burden of high disease activity and its association with damage accrual and clinical outcomes in patients with cSLE.MethodsThis retrospective cohort study included cSLE patients with at least one year of follow-up. High disease activity status (HDAS) was defined as a SLEDAI-2 K score ≥10. Patients were classified into two groups: HDAS (ever experienced HDAS) and non-HDAS. Time-adjusted cumulative HDAS (cHDAS) was defined as the proportion of time spent in HDAS during the observation period. Clinical features, laboratory parameters, treatment, organ damage assessed by the SLICC/ACR Damage Index, and remission status were recorded at baseline and each follow-up visit. Logistic regression, Cox proportional hazards models, and Kaplan–Meier analyses were performed.ResultsAmong 196 cSLE (mean age at diagnosis 11.3 ± 2.9 years; 82.1% female), 126 (64.3%) experienced high disease activity status (HDAS). Patients with HDAS had a significantly higher rate of damage accrual than those without (SDI ≥ 1: 33.3% vs. 11.4%, p < 0.001), with an odds ratio of 3.0 (95% CI 1.2–7.3; p = 0.017). Time-adjusted cHDAS was significantly higher in patients with damage than in those without (11.8 [2.4–33.0] vs. 1.5 [0–5.4], p < 0.001). Increasing thresholds of time-adjusted cHDAS were associated with a progressively higher risk of damage accrual, with hazard ratios rising from 6.2 at ≥5% to 16.1 at ≥15% (all p < 0.001). The mean time to damage was significantly shorter in patients with time-adjusted cHDAS ≥5% compared with <5% (5.0 [3.7–6.2] vs. 12.5 [11.3–13.7] years, p < 0.001). In multivariable analysis, neuropsychiatric involvement and time-adjusted cHDAS ≥5% remained independent predictors of damage accrual, with hazard ratios of 2.4 (95% CI 1.3–4.4, p = 0.008) and 4.2 (95% CI 2.1–8.3, p < 0.001), respectively.ConclusionsHigher time-adjusted cHDAS was associated with an increased risk of damage accrual, indicating that greater cumulative exposure to high disease activity is associated with damage. Neuropsychiatric involvement and time-adjusted cHDAS ≥5% were significantly associated with damage accrual. Introduction Systemic lupus erythematosus (SLE) is a systemic autoimmune disease affecting multiple organ systems, and it has diverse clinical manifestations. SLE can lead to significant morbidity and mortality. Clinical manifestations and disease severity vary among patients based on the affected organ systems. Childhood-onset SLE (cSLE) is defined as disease onset before the age of 18 years. cSLE affects approximately 15%–20% of the overall SLE population, with a prevalence rate of 3.3–8.8 per 100,000 children (1). The average age at diagnosis is 11–12 years, and onset before the age of 5 years is relatively rare. The etiological, pathogenic, and laboratory features of cSLE are similar to those of adult-onset SLE. However, the two conditions differ in terms of the frequency and severity of clinical manifestations. Notably, cSLE is more likely to present with a more severe disease course, a higher frequency of renal and neuropsychiatric involvement, and poorer outcomes compared with adult SLE (2). In SLE, assessing disease activity is essential for detecting flare-ups, monitoring therapeutic efficacy, and predicting clinical outcomes. There are several available tools for evaluating disease severity. The Systemic Lupus Erythematosus Disease Activity Index 2000 (SLEDAI-2K) is one of the most widely used indices in clinical research. Previous studies have shown that patients with a high disease activity status (HDAS) are at increased risk for severe organ involvement, particularly neuropsychiatric manifestations, renal involvement, and vasculitis. These patients are more likely to have low baseline complement levels and test positive for anti-double-stranded DNA (anti-dsDNA) and anti-Smith antibodies. They also require intensive immunosuppressive medications and are at increased risk of damage accrual and disease flare-ups (3). In addition, recurrent or prolonged periods of HDAS have been associated with younger age at disease onset, more severe clinical presentations, a greater serological activity, and a higher risk of damage accrual. A total cumulative duration of HDAS (cHDAS) of >2 years can significantly affect long-term organ damage (4). In another study, patients with an extremely high initial disease activity are less likely to achieve a low disease activity state during the disease course (5). Despite these findings, the existing literature is largely derived from adult SLE cohorts, while studies focusing on cSLE remain limited. Furthermore, most prior research has emphasized low disease activity states and their association with improved outcomes, whereas relatively less attention has been given to the impact of high disease activity on the risk of damage. In particular, the effects of the duration and cumulative burden of high disease activity on damage accrual in cSLE have not been well established. Therefore, this study aimed to evaluate whether high disease activity, including its cumulative duration, is associated with adverse disease outcomes, particularly damage accrual, in patients with cSLE. Materials and methods This retrospective single-center cohort study included all patients who had been diagnosed with SLE according to the 1997 American College of Rheumatology (ACR) criteria (6), the 2012 Systemic Lupus International Collaborating Clinics criteria (7), or the 2019 European League Against Rheumatism/ACR Classification Criteria for Systemic Lupus Erythematosus (8) before the age of 18 years, and who had at least 1 year of follow-up at Ramathibodi Hospital, a tertiary referral center, between January 2008 and March 2023. Observation time was defined as the duration from cohort entry (date of diagnosis or first eligible visit at the study center) to the occurrence of disease damage or the last follow-up visit, whichever occurred first; this corresponded to time to damage for patients who developed damage and total follow-up time for those who were censored. Patients were managed through a multidisciplinary care approach involving pediatric rheumatologists and nephrologists, with routine standardized assessment of disease activity and management. Patients with incomplete medical records or a diagnosis of overlap syndrome were excluded from the study. The current study was approved by the Ethics Committee of the Faculty of Medicine, Ramathibodi Hospital, Mahidol University (MURA2024/214), and the need for informed consent was waived due to the retrospective nature of the study. Data collection Baseline characteristics The data of eligible patients were extracted from the medical records. Details of the demographic and baseline characteristics of the participants, including age, age at diagnosis, sex, disease duration, and duration prior to treatment initiation, were collected. The clinical manifestations included initial presenting symptoms, SLEDAI-2K score at diagnosis, and relevant laboratory findings such as complete blood count, erythrocyte sedimentation rate, complement levels, antinuclear antibody, anti-dsDNA, direct Coombs test, and antiphospholipid antibodies. Lupus nephritis (LN) classification was based on renal biopsy findings. Treatment information, including the use of medications such as hydroxychloroquine, azathioprine, methotrexate, mycophenolate mofetil, and rituximab, was recorded. Cumulative doses of prednisolone and cyclophosphamide were calculated as the sum of all administered doses over the observation period. In addition, the time-adjusted cumulative prednisolone dose was calculated by dividing the total prednisolone-equivalent dose by the total observation time. Corticosteroid doses were converted to prednisolone-equivalent doses prior to calculation. Infectious events, particularly disseminated infections, those involving major organs, or those requiring hospitalization, were documented. Data on disease flare-ups and remissions were also collected during the follow-up period. Disease activity assessment Disease activity was assessed using the SLEDAI-2K, which included 24 important clinical and laboratory manifestations, with a total score of 0–105 (9). Time-adjusted mean SLEDAI-2K was used to assess disease activity over time, reflecting the average disease activity throughout follow-up. It was calculated as the area under the curve (AUC) of SLEDAI-2K over time, divided by the total duration of follow-up. The AUC was estimated using the trapezoidal rule (supplementary material) (10). The Physician Global Assessment (PGA), with a clinician-rated scale of 0–3 (11). The SLEDAI-2K and PGA scores were recorded at baseline, every 3–6 months, and at each visit during active disease or flares until the most recent follow-up visit. HDAS was defined as a SLEDAI-2K score ≥10 (3, 4). Patients were classified into two groups based on whether they ever experienced HDAS during follow-up (i.e., ≥1 visit with SLEDAI-2K ≥10), as the HDAS group and the non-HDAS group. Cumulative HDAS (cHDAS) was defined as the total duration of time during which patients were in HDAS over the observation period. The duration of HDAS was estimated by summing time intervals between consecutive visits with SLEDAI-2K ≥10. For intervals in which SLEDAI-2K crossed the HDAS threshold between two visits, the transition time was estimated using linear interpolation, assuming a linear change between visits. To account for differences in follow-up duration across patients, time-adjusted cHDAS was calculated as the proportion of cHDAS divided by the total observation time. Time-adjusted cHDAS was used as the primary exposure variable in all analyses. The calculation formula is shown in the supplementary material. Outcome assessment Organ damage Organ damage was evaluated using the 2012 Systemic Lupus International Collaborating Clinics/ACR Damage Index (SDI) developed in 1996. The SDI included 12 organ systems and assigned scores based on irreversible damage that had been present for at least 6 months, whether due to disease activity, treatment, or complications, with a maximum total score of 47 (12). The patients in this study received routine eye exams every 6–12 months. Disease flare-up Flare-ups were classified as mild-to-moderate or severe. A mild-to-moderate flare-up was defined by an increase in the SLEDAI-2K score to ≥3; new or worsening involvement of the skin, mucous membranes, serositis, arthritis, or fever; an increase in PGA score to 1–2.5; a slight increase in prednisolone dose (< 0.5 mg/kg/day), or addition of nonsteroidal anti-inflammatory drugs or hydroxychloroquine. A severe flare-up was defined by an increase in the SLEDAI-2K score to >12; new or worsening neuropsychiatric symptoms, vasculitis, nephritis, myositis, thrombocytopenia (platelet count <60,000/μL), and hemolytic anemia with hemoglobin levels <7 g/dL; a significant increase in prednisolone dosage (>2 times at baseline or >0.5 mg/kg/day); hospitalization due to disease activity; addition of a new immunosuppressive agent; or a PGA score of >2.5 (13). Low disease activity Low disease activity was examined using three definitions: minimal disease activity (MDA), low disease activity (LDA), and lupus low disease activity state (LLDAS). MDA was defined as a clinical SLEDAI-2K score ≤1 (excluding serologic activity such as anti-dsDNA positivity and/or low complement levels), maintained with antimalarials, standard immunosuppressive doses (not exceeding the recommended doses), and low-dose prednisolone (≤5 mg/day). Only hematologic abnormalities directly attributed to lupus, such as isolated leukopenia and thrombocytopenia, were allowed under this definition (14). LDA was defined as SLEDAI-2K scores ≤2, with or without serologic activity, and limited to the presence of only one of the following conditions: rash, alopecia, mucosal ulcers, pleurisy, pericarditis, fever, thrombocytopenia, or leukopenia. Patients meeting the LDA criteria were treated with antimalarials only, without glucocorticoids or immunosuppressants (15). LLDAS was defined as a SLEDAI-2K score ≤4, with no active renal, central nervous system, cardiopulmonary, gastrointestinal, or vascular involvement; no fever or hemolytic anemia; and no new lupus disease activity compared with the previous visit. It also required a PGA score of ≤1, current treatment with prednisolone (or equivalent) at a dose of ≤7.5 mg/day, and well-tolerated standard maintenance doses of immunosuppressive and/or approved biologic agents, excluding investigational drugs (16). Disease remission Disease remission was categorized into remission on therapy, remission off therapy, and complete remission. Remission on therapy was defined as a clinical SLEDAI-2K score of 0, a PGA score of <0.5, a prednisolone dose of ≤5 mg/day or ≤0.15 mg/kg/day, and the absence of escalation or the addition of immunosuppressive or biologic therapy. Remission off therapy was defined as clinical remission without systemic treatments for SLE, other than maintenance antimalarial therapy. Complete remission was defined as the achievement of clinical and serological remission (normal anti-dsDNA and complement levels) without the use of immunosuppressive agents or glucocorticoids, other than ongoing antimalarial treatment (17, 18). Statistical analysis Descriptive statistics were reported as mean (standard deviation, SD) or median (interquartile range, IQR). The Mann–Whitney U-test and the chi-square test were used to compare between-group differences. Flare rates were analyzed using Poisson regression models with a log link function, with the number of flares as the dependent variable and follow-up time included as an offset term to account for varying observation periods. Incidence rate ratios (IRRs) with 95% confidence intervals were reported to compare flare rates between groups. Logistic regression analysis was performed to identify factors associated with ever experiencing HDAS. Candidate variables were selected based on clinical relevance and evidence from prior literature, including demographic characteristics (age, sex, disease duration), clinical factors (damage accrual), and treatment-related variables (e.g., cumulative glucocorticoid exposure and history of infection). Variables that are components of the SLEDAI-2K score were not included in the model to avoid circularity. Kaplan–Meier survival analysis and Cox proportional hazards models were conducted to evaluate factors affecting the occurrence of damage accrual. Variables included in the Cox proportional hazards model were selected based on clinical relevance and factors associated with damage accrual in univariable analyses. The number of covariates included in the model was limited according to the number of outcome events to avoid overfitting. In addition, variables with high collinearity in logistic regression and the Cox proportional hazards model were excluded based on correlation analysis or clinical judgment. All statistical analyses were performed using the Statistical Package for the Social Sciences software version 29.0.1.0. (IBM Corp., Armonk, NY, the USA). A p-value of <0.05 was considered statistically significant. Results In total, 230 patients were diagnosed with cSLE, and 34 patients with incomplete medical histories were excluded from the analysis. Of the 196 patients with SLE in this study, 161 (82.1%) were female. The mean age at diagnosis was 11.3 ± 2.9 years, the median disease duration was 3.4 (IQR 1.8–6.1) years, and the median duration before diagnosis was 1.2 (IQR 0.5–3.2) months. The most common manifestations were hematologic involvement (71.4%) and mucocutaneous organ involvement (62.2%). The other clinical features commonly found in this cohort included renal condition, fever, vasculitis, serositis, and neurological involvement. The median baseline SLEDAI-2K and time-adjusted mean SLEDAI-2K score were 10.5 (IQR 6.0–17.0) and 2.0 (IQR 0.8–3.4). A total of 126 (64.3%) patients had HDAS. The median time-adjusted cHDAS was 6.1 (IQR 2.8–12.5). The HDAS and non-HDAS groups did not differ significantly in sex, age, or disease duration. Almost all patients with neurological involvement had HDAS. In addition, the HDAS group had a significantly higher frequency of renal involvement, particularly LN class III-V, serositis, vasculitis, mucocutaneous and musculoskeletal involvement, compared with the non-HDAS group. In terms of laboratory results, the HDAS group had significantly lower white blood cell counts, absolute lymphocyte counts, complement levels, and a higher erythrocyte sedimentation rate, and was more likely to present with anti-dsDNA antibody positivity than the non-HDAS group. Almost all patients in both groups were treated with prednisolone (99.5%) and hydroxychloroquine (95.9%), with no significant differences between the two groups. In addition, no chloroquine was used in this study. However, the HDAS group had a higher frequency of use of azathioprine, mycophenolate mofetil, and cyclophosphamide. Moreover, they received a higher median cumulative dose of prednisolone and cyclophosphamide than the non-HDAS group (Table 1). We further evaluated factors associated with HDAS and found that a time-adjusted cumulative prednisolone dose >0.15 mg/kg/day and the presence of damage accrual were independently associated with HDAS (Table 2). Table 1Baseline characteristics of cSLE patients with and without high disease activity.CharacteristicsAll patientsHDASNon-HDASp-value(n = 196)(n = 126)(n = 70)Female, n (%)161 (82.1)103 (81.7)58 (82.9)0.846Age at diagnosisa11.3 ± 2.911.4 ± 2.611.1 ± 3.30.480Duration of disease (years)3.4 (1.8–6.1)3.0 (1.5–5.6)4.2 (2.4–6.6)0.093Duration before diagnosis (months)1.2 (0.5–3.2)1.0 (0.5–2.3)1.4 (0.6–3.6)0.036*Number of visits17 (12–26)16 (11–24)18 (13–31)0.143Clinical manifestation, n (%)- Fever51 (26.0)35 (27.8)16 (22.9)0.50- Hematological140 (71.4)91 (70.0)49 (72.2)0.744- Mucocutaneous122 (62.2)85 (67.5)37 (52.9)0.047*- Neurological31 (15.4)29 (23.0)1 (1.4)<0.001*- Serositis33 (16.8)31 (24.6)2 (2.9)<0.001*- Vasculitis31 (15.8)28 (22.2)3 (4.3)<0.001*- Lupus nephritisb106 (54.1)93 (73.8)13 (18.6)<0.001*- Lupus nephritis class III-V71 (36.2)65 (51.6)6 (8.6)<0.001*- Musculoskeletal69 (35.2)49 (38.9)15 (21.4)0.017*- Gastrointestinal11 (5.6)8 (6.3)3 (4.3)0.749SLEDAI-2K at diagnosis10.5 (6.0–17.0)15.0 (12.0–19.0)6.0 (4.0–8.0)<0.001*Time-adjusted mean SLEDAI-2K2.0 (0.8–3.4)2.4 (1.3–4.0)1.0 (0.2–2.0)<0.001*LaboratoryHematocrita30.6 ± 7.030.7 ± 5.830.2 ± 8.90.866WBC (×103 cells/mm3)5.4 (3.7–8.3)5.0 (3.3–7.2)6.0 (4.7–8.9)<0.002*Absolute lymphocyte count (×103 cells/mm3)1.6 (1.1–2.4)1.3 (0.9–2.0)1.9 (1.3–2.8)<0.001*Leukopenia/lymphopenia, n (%)84 (45.7)64 (54.7)20 (29.9)0.001*Platelet (×103 cells/mm3)219 (137–311)214 (132–297)231 (99–336)0.477Thrombocytopenia, n (%)45 (24.1)26 (21.7)19 (28.4)0.305Erythrocyte sedimentation rate (mm/h)60 (28–87)65 (35.5–88)53.5 (22–77.5)0.051C3 (g/L)0.57 (0.31–0.95)0.40 (0.27–0.74)0.95 (0.65–1.24)<0.001*C4 (g/L)0.07 (0.03–0.16)0.06 (0.03–0.10)0.16 (0.07–0.27)<0.001*Hypocomplementemia, n (%)136 (73.9)105 (89.0)31 (47.0)<0.001*Anti-dsDNA, n (%)134 (74.0)104 (90.4)30 (45.5)<0.001*Direct Coombs' test, n (%)94 (62.3)63 (64.9)31 (57.4)0.360Antiphospholipid antibodies, n (%)26 (16.7)18 (18.9)8 (13.1)0.340TreatmentHydroxychloroquine, n (%)188 (95.9)120 (95.2)68 (97.1)0.714Prednisolone, n (%)195 (99.5)126 (100.0)69 (98.6)0.357Cumulative dose of prednisolone (mg/kg)265.6 (171–459)317.2 (199.9–487.0)220.2 (136.1–343.6)<0.001*Time-adjusted cumulative prednisolone dose0.26 (0.15–0.48)0.29 (0.21–0.53)0.17 (0.07–0.29)<0.001*Cyclophosphamide, n (%)91 (46.4)83 (65.9)8 (11.4)<0.001*Cumulative dose of cyclophosphamide (mg/m2)0 (0–5,665)4,313 (0–6,140.3)0 (0–0)<0.001*Mycophenolate mofetil, n (%)46 (23.5)33 (26.2)13 (18.6)0.228Azathioprine, n (%)139 (70.9)96 (76.2)43 (61.4)0.029*Methotrexate, n (%)12 (6.1)8 (6.3)4 (5.7)1.00Cyclosporine, n (%)5 (2.6)2 (1.6)3 (4.3)0.351Immunoglobulin, n (%)9 (4.6)7 (5.6)2 (2.9)0.495Rituximab, n (%)4 (2.0)2 (1.6)2 (2.9)0.618Data was presented as median (P25–75);.amean ± SD.*p < 0.05 was set as significance.HDAS, High disease activity status; SLEDAI-2K, The Systemic Lupus Erythematosus Disease Activity Index 2000; WBC, white blood cell; C, complement component; Anti-dsDNA, anti-double stranded DNA.bAmong patients with lupus nephritis, renal biopsy was performed in 74/93 (79.6%) in the HDAS group and 9/13 (69.2%) in the non-HDAS group. Baseline characteristics of cSLE patients with and without high disease activity. Data was presented as median (P25–75);. mean ± SD. p < 0.05 was set as significance. HDAS, High disease activity status; SLEDAI-2K, The Systemic Lupus Erythematosus Disease Activity Index 2000; WBC, white blood cell; C, complement component; Anti-dsDNA, anti-double stranded DNA. Among patients with lupus nephritis, renal biopsy was performed in 74/93 (79.6%) in the HDAS group and 9/13 (69.2%) in the non-HDAS group. Table 2Factors associated with ever experiencing high disease activity status: logistic regression analysis.FactorsUnivariable analysisMultivariable analysisOR (95%CI)p-valueOR (95%CI)p-valueGender0.9 (0.4–2.0)0.8460.9 (0.4–2.0)0.718Age at diagnosis (year)1.0 (0.9–1.1)0.4791.1 (1.0–1.2)0.121Disease duration0.9 (0.9–1.0)0.1851.1 (1.0–1.2)0.181Time-adjusted cumulative prednisolone dose >0.15 mg/kg/day3.8 (1.9–7.4)<0.001*4.3 (1.9–9.7)<0.001*Damage accrual1.9 (1.3–2.7)<0.001*3.0 (1.2–7.3)0.017*Infection2.9 (1.4–5.9)0.004*2.0 (0.9–4.4)0.072*p < 0.05. Factors associated with ever experiencing high disease activity status: logistic regression analysis. p < 0.05. Disease outcomes Among the 196 patients with cSLE, 50 (25.5%) developed organ damage. The most frequently affected systems were ocular and neuropsychiatric, followed by musculoskeletal damage. Cataract accounted for 81.3% of all ocular complications, and the frequency of glaucoma (12.5%), retinal scar (6.3%), and central scotoma (6.3%) was lower. Neuropsychiatric involvement comprised seizures requiring prolonged therapy (43.8%), cerebrovascular accidents (37.5%), and cognitive impairment or major psychosis (37.5%) (Supplementary Table S1). In our cohort, all patients had no end-organ damage at baseline. Damage accrual was more common in the HDAS group than in the non-HDAS group (42 [33.3%] vs. 8 [11.4%], p = <0.001). Ocular damage occurred significantly more often in the HDAS group than in the non-HDAS group (14 [11.1%] vs. 2 [2.9%], p = 0.043) (Table 3). Then, we assessed whether high disease activity was associated with damage accrual by using time-adjusted cHDAS. The median time-adjusted cHDAS differed significantly between the SDI and non-SDI groups (11.8 (IQR 2.4–33.0) vs. 1.5 (IQR 0–5.4), p < 0.001). Moreover, higher time-adjusted cHDAS thresholds were consistently associated with an increased risk of damage accrual (SDI ≥ 1). Specifically, patients with time-adjusted cHDAS ≥ 5% had a 6.2-fold higher risk of damage accrual, which increased progressively to 8.4-fold at ≥7.5%, 12.2-fold at ≥10%, and 16.1-fold at ≥15%. All associations were statistically significant (p < 0.001; see Table 4). Furthermore, the mean time to damage accrual was significantly shorter in patients with time-adjusted cHDAS ≥5% than in those with time-adjusted cHDAS <5% (5.0 years (3.7–6.2) vs. 12.5 years (11.3–13.7), p < 0.001) (Figure 1). Similarly, higher percentages of time-adjusted cHDAS were associated with a shorter time to damage accrual (Table 5). Table 3Disease outcomes of cSLE patients with and without high disease activity.Disease damageAll patientsHDASNon-HDASp-value(n = 196)(n = 126)(n = 70)SDI, n (%)50 (25.5)42 (33.3)8 (11.4)<0.001*- Ocular16 (8.2)14 (11.1)2 (2.9)0.043*- Neuropsychiatric16 (8.2)13 (10.3)3 (4.3)0.178- Renal2 (1.0)2 (1.6)00.538- Pulmonary3 (1.5)3 (2.4)0 (0)0.554- Cardiovascular000-- Peripheral vascular1 (0.5)1 (0.8)01- Musculoskeletal14 (7.1)12 (9.5)2 (2.9)0.082- Skin2 (1.0)02 (2.9)0.126- Gastrointestinal000-- Gonadal failure1 (0.5)1 (0.8)01- Diabetes mellitus5 (2.6)5 (4.0)00.162- Malignancy1 (0.5)1 (0.8)01Infection, n (%)59 (30.1)47 (37.3)12 (17.1)0.003*Flare, n (%)78 (39.8)48 (38.1)30 (42.9)0.514Flare rate (per 100 person-years)17.116.817.6–Incidence rate ratio for total flaresa0.95 (95% CI 0.68–1.33)0.772Mild to moderate flare rate (per 100 person-year)11.39.214.5-Incidence rate ratio for mild to moderate flaresa0.63 (95% CI 0.42–0.95)0.027*Severe flare rate (per 100 person-year)5.87.63.1-Incidence rate ratio for severe flaresa2.45 (95% CI 1.22–4.93)0.012*Duration before flare (years)3.0 (1.4–5.1)3.0 (1.5–5.0)2.8 (0.9–5.6)0.606Low disease activity and remission, n (%)LLDAS141 (71.9)84 (66.7)57 (81.4)0.028*LDA43 (21.9)16 (12.7)27 (38.6)<0.001*MDA142 (72.4)84 (66.7)58 (82.9)0.015*Remission on medication123 (62.8)75 (59.5)48 (68.6)0.209Remission off medication38 (19.4)14 (11.1)24 (34.3)<0.001*Complete remission23 (11.7)7 (5.6)16 (22.9)<0.001**p < 0.05 was set as significance,.SDI, Systemic Lupus International Collaborating Clinics/American College of Rheumatology (SLICC/ACR) Damage Index; MDA, minimal disease activity; LDA, low disease activity; LLDAS, lupus low disease activity.aIncidence rata ratio for total flares, mild to moderate flares, and severe flares using the non-HDAS group as a reference. Disease outcomes of cSLE patients with and without high disease activity. p < 0.05 was set as significance,. SDI, Systemic Lupus International Collaborating Clinics/American College of Rheumatology (SLICC/ACR) Damage Index; MDA, minimal disease activity; LDA, low disease activity; LLDAS, lupus low disease activity. Incidence rata ratio for total flares, mild to moderate flares, and severe flares using the non-HDAS group as a reference. Table 4The association between time-adjusted cumulative high disease activity (cHDAS) and damage accrual (SDI ≥ 1).Time-adjusted cHDASHazard Ratio (95%CI)p-value≥5%6.2 (3.4–11.3)<0.001*≥7.5%8.4 (4.6–15.3)<0.001*≥10%12.2 (6.6–22.6)<0.001*≥15%16.1 (8.4–30.9)<0.001**A p < 0.05 was considered statistically significant.SDI, Systemic Lupus International Collaborating Clinics/American College of Rheumatology Damage Index; CI, confidence interval. The association between time-adjusted cumulative high disease activity (cHDAS) and damage accrual (SDI ≥ 1). A p < 0.05 was considered statistically significant. SDI, Systemic Lupus International Collaborating Clinics/American College of Rheumatology Damage Index; CI, confidence interval. Kaplan–meier analysis of time to disease damage in cSLE patients stratified by time-adjusted cumulative high disease activity status (cHDAS) (≥5% vs. <5%). Table 5The kaplan–meier analysis related to the percentage of time-adjusted cHDAS and the development of damage accrual (SDI ≥ 1).% Time-adjusted cHDASMean time to disease damage (years)Median time to disease damage (years)p-value≥5%5.0 (95% CI 3.7–6.2)4.6 (95% CI 1.5–7.7)<0.001*≥7.5%3.9 (95% CI 2.6–5.1)2.2 (95% CI 1.6–2.8)<0.001*≥10%2.8 (95% CI 1.8–3.8)1.5 (95% CI 0.4–2.6)<0.001*≥15%1.5 (95% CI 0.8–2.2)0.6 (95% CI 0.3–0.8)<0.001**p < 0.05 was set as significance; cHDAS, cumulative high disease activity status; Time-adjusted cHDAS, cHDAS divided by disease duration; SDI, Systemic Lupus International Collaborating Clinics/American College of Rheumatology Damage Index. The kaplan–meier analysis related to the percentage of time-adjusted cHDAS and the development of damage accrual (SDI ≥ 1). p < 0.05 was set as significance; cHDAS, cumulative high disease activity status; Time-adjusted cHDAS, cHDAS divided by disease duration; SDI, Systemic Lupus International Collaborating Clinics/American College of Rheumatology Damage Index. Among the patients in this cohort, 141 (71.9%) achieved low disease activity as defined by LLDAS. Additionally, 123 (62.8%) achieved remission on medication, while 38 (19.4%) achieved remission off medication; 23 patients (11.7%) met criteria for complete remission. Infections occurred in 59 patients (30.1%), with a higher proportion observed in the HDAS group compared with the non-HDAS group (37.3% vs. 17.1%, p = 0.003). Overall, 78 patients (39.8%) experienced disease flares. The overall flare rate was similar between groups (16.8 vs. 17.6 per 100 person-years; IRR 0.95, 95% CI 0.68–1.33, p = 0.772). However, when stratified by severity, the rate of mild-to-moderate flares was lower in the HDAS group (9.2 vs. 14.5 per 100 person-years; IRR 0.63, 95% CI 0.42–0.95, p = 0.027), whereas the rate of severe flares was higher (7.6 vs. 3.1 per 100 person-years; IRR 2.45, 95% CI 1.22–4.93, p = 0.012), as shown in Table 3. Predictors of damage accrual in patients with childhood-onset SLE In the multivariable analysis, after adjustment for confounders, neuropsychiatric SLE (NPSLE) and time-adjusted cHDAS ≥5% remained independent predictors of damage accrual, with hazard ratios of 2.4 (95% CI 1.3–4.4, p = 0.008) and 4.2 (95% CI 2.1–8.3, p < 0.001), respectively. In contrast, time-adjusted cumulative prednisolone dose ≤0.15 mg/kg/day was no longer significantly associated with damage accrual (HR 0.5, 95% CI 0.2–1.3, p = 0.167) (Table 6). To further delineate the independent effects of glucocorticoid exposure from disease activity, additional analyses were performed to evaluate the association between time-adjusted cumulative prednisolone dose and damage accrual separately from HDAS. In these analyses, a higher time-adjusted cumulative prednisolone dose was associated with a shorter time to damage accrual in Kaplan–Meier analysis, and with an increased risk of damage accrual in univariable analyses (Supplementary Tables S2, S3). Table 6Factors associated with damage accrual: Cox proportional hazards analysis.FactorsUnivariable analysisMultivariable analysisHR (95%CI)p-valueHR (95%CI)p-valueAge at diagnosis1.0 (0.9–1.1)0.9241.0 (0.9–1.1)0.751Female1.1 (0.5–2.3)0.8361.4 (0.6–3.0)0.418NPSLE3.2 (1.8–5.9)<0.001*2.4 (1.3–4.4)0.008*Time-adjusted cHDAS ≥5%6.2 (3.4–11.3)<0.001*4.2 (2.1–8.3)<0.001*Time-adjusted cumulative prednisolone dose ≤0.15 mg/kg/day0.3 (0.1–0.6)0.002*0.5 (0.2–1.3)0.167*p < 0.05 was set as significance.cHDAS, cumulative high disease activity status; Time-adjusted cHDAS, cHDAS divided by disease duration. HR, hazard ratio; CI, confidence interval. Factors associated with damage accrual: Cox proportional hazards analysis. p < 0.05 was set as significance. cHDAS, cumulative high disease activity status; Time-adjusted cHDAS, cHDAS divided by disease duration. HR, hazard ratio; CI, confidence interval. Discussion Understanding the impact of disease activity on disease outcomes is essential in the management of cSLE. Persistent high disease activity and treatment-related adverse effects may contribute to irreversible organ damage; therefore, evaluating the role of HDAS, particularly its cumulative burden, is critical for optimizing long-term care. In this study, patients with HDAS had more severe disease phenotypes, greater cumulative glucocorticoid exposure, and were less likely to achieve low disease activity or remission, resulting in poorer outcomes, including infections and severe flares. Furthermore, increasing levels of time-adjusted cHDAS were associated with a progressively higher risk of damage accrual and a shorter time to damage. In multivariable analysis, time-adjusted cHDAS ≥ 5% and NPSLE were identified as independent predictors of damage accrual. These findings suggest that cumulative burden of high disease activity plays a critical role in disease outcomes in cSLE. In this study, we defined HDAS as a SLEDAI-2K score ≥10, based on previous literature demonstrating that this threshold identifies patients with more severe disease, greater treatment burden, and a higher risk of damage accrual (3, 4, 19, 20). To assess the impact of high disease activity on outcomes, patients were classified according to whether they had ever experienced HDAS during the observation period, rather than relying only on baseline disease activity, thereby better reflecting the longitudinal disease course. Given that cumulative disease activity over time is a strong predictor of damage accrual (21), we extended this concept to cumulative high disease activity. To quantify this burden, we used time-adjusted cHDAS, defined as the proportion of follow-up time spent in HDAS. This approach provides a more comprehensive assessment of sustained high disease activity and, importantly, normalizes for differences in follow-up duration across patients, enabling more reliable comparisons. Our findings are consistent with those of previous studies (3, 4), which showed that patients with HDAS were more likely to present with NPSLE, LN, serositis, and that positive anti-dsDNA antibodies and low complement levels were also associated with HDAS. Although these studies were conducted primarily on adult populations, the associations were similar in our pediatric cohort. This is biologically plausible, as patients with more severe lupus were more likely to test positive for anti-dsDNA and hypocomplementemia, which reflect a high disease activity (3). Although the proportion of female patients in our cohort was slightly lower than that reported in some SLE cohorts, it remained within the expected range for cSLE, where females account for approximately 75%–90% of cases, with increasing female predominance during adolescence (22–26). The pattern of organ involvement was comparable to that reported in previous studies, with hematologic, mucocutaneous, and musculoskeletal manifestations being the most common, followed by LN (approximately 40%) and NPSLE (approximately 10%–15%) (22, 23, 26). In this study, approximately two-thirds of patients in this cohort had HDAS. The main factors associated with HDAS were time-adjusted cumulative prednisolone dose >0.15 mg/kg/day and presence of damage accrual. During periods of HDAS, patients more frequently exhibited severe clinical features, including NPSLE, renal involvement (especially LN class III–V), serositis, and vasculitis. This higher disease burden likely required higher cumulative prednisolone use and may have contributed to subsequent damage accrual. The most frequent damage to the system/organ in patients with HDAS in this study differed from that reported in previous ones (3). Based on an earlier study, the main associated with HDAS was renal damage accrual. Meanwhile, our findings showed that the predominant damage accrual was ocular complications, and there was no significant association with renal outcomes. This discrepancy may be explained by differences in study populations. In a previous study (3), most patients were adults, with approximately half of them being older than 45 years. In elderly patients, LN often developed >5 years after diagnosis, and it was compounded by comorbidities such as hypertension, leading to chronic renal damage, and was more challenging to manage (27). In contrast, most children with cSLE present with LN at diagnosis (28, 29) and typically receive early and aggressive treatment, which may prevent long-term renal damage. Ocular damage is likely related to corticosteroid use. This is because corticosteroids alter the biological structure of the trabecular meshwork and increase intraocular pressure (30–32). Moreover, children are particularly susceptible because their trabecular meshwork is immature (33). Thus, they are vulnerable to steroid-induced ocular complications even at lower doses and shorter treatment durations compared with adults (34). Although our results demonstrated that HDAS was associated with damage accrual, the cumulative burden and timing of cHDAS had not been fully explored. Therefore, we further evaluated multiple time-adjusted cHDAS thresholds as predictors of damage accrual, while adjusting for potential confounders in a multivariable analysis. We found that the presence of NPSLE and time-adjusted cHDAS ≥5% were independently associated with an increased risk of damage accrual. These findings are consistent with previously identified risk factors for damage accrual (3, 4, 21, 35–37). However, most prior studies have quantified HDAS based on the number of episodes rather than its cumulative burden over time. In contrast, our approach incorporates the longitudinal burden of HDAS, which may better reflect the overall disease course in patients with cSLE. Patients with NPSLE had a higher burden, as reflected in the SLEDAI-2K score, which is also related to HDAS. NPSLE commonly presents with psychosis and seizures, which reasonably account for complications such as cognitive impairment, seizure requiring prolonged therapy, and cerebrovascular accidents. Antiphospholipid antibodies can increase the risk of venous and arterial thrombosis (38), and a previous study showed that they play a role in thrombotic damage (39), particularly in NPSLE. However, in our cohort, only three patients with NPSLE tested positive for antiphospholipid antibodies, of whom one presented with thrombosis. The patients who tested negative for these antibodies also developed cerebrovascular accidents. This finding indicated that the role of antiphospholipid antibodies in organ damage among patients with NPSLE remains controversial, as thrombotic events in SLE are influenced by multiple factors, not only antiphospholipid antibodies. In addition to high disease activity and neurological involvement, which were associated with damage accrual in this study, a previous study had also highlighted severe disease flares as a factor associated with increased damage accrual (40). Although our patients with HDAS experienced a significantly higher incidence of severe flares compared with those without HDAS, severe flares were not significantly associated with damage accrual in univariable analysis. This suggests that, while severe flares are more frequent in patients with high disease activity, they may not independently predict damage accrual in our cohort. The difference between the previous study and this study may reflect variations in analytical approach. A previous study primarily used Chi-square or Fisher's exact tests, whereas we applied Cox proportional hazards analysis to account for time-to-event data; however, as the association was not significant in univariable analysis, it was not carried forward into the multivariable model. Corticosteroids remain indispensable in the management of cSLE; they are also a major contributor to organ damage (21, 41, 42), including cataracts, avascular necrosis, osteoporotic fractures, and coronary artery disease (43). In this study, the time-adjusted cumulative prednisolone dose was statistically significant in univariable analysis but not in the multivariable model. This attenuation may be explained by confounding, as patients with higher disease activity are more likely to receive greater cumulative corticosteroid exposure. To further explore this relationship, we conducted an additional analysis examining the effect of increasing cumulative prednisolone dose on damage accrual. We observed a dose–response relationship, in which higher time-adjusted cumulative prednisolone doses were associated with both a shorter time to, and an increased risk of, damage accrual. Approximately 7% of patients developed musculoskeletal damage, most commonly avascular necrosis, followed by osteoporotic fractures and joint deformities. The median time to avascular necrosis was 2.7 years, which is shorter than that reported in adult-onset SLE (aSLE) studies (44–50). A likely explanation is that our patients were younger at disease onset, a known risk factor for avascular necrosis in SLE (51–53). Based on previous studies early-onset avascular necrosis can occur in patients with SLE even with a lower cumulative corticosteroid exposure. These patients develop avascular necrosis more frequently compared with those with other rheumatic diseases (54, 55). This finding may be explained by the disease itself. This is because SLE is associated with vascular involvement and an increased risk of thrombosis due to alterations in coagulation, particularly in patients with LN (56, 57). Previous studies have demonstrated that longer disease duration is associated with an increased risk of organ damage (35, 36, 58), and many have emphasized that achieving LLDAS is associated with reduced long-term damage (59–64). Building on this concept, we propose that the cumulative burden of high disease activity should also be considered, as failure to promptly control high disease activity may independently contribute to damage accrual. Consistent with this hypothesis, this study showed that patients with cSLE who had a time-adjusted cHDAS ≥5% had approximately a 4-fold higher risk of damage accrual compared with those with a time-adjusted cHDAS <5%. This study has several limitations. It included a relatively small sample size, employed a retrospective design, and had a moderate follow-up duration, which may have led to an underestimation of damage accrual. In addition, it was conducted at a single tertiary care center, potentially limiting data completeness and generalizability. Therefore, prospective multicenter studies are warranted to validate these findings. Despite these limitations, the study has notable strengths. The longitudinal assessment of disease activity and treatment exposure using time-adjusted measures, including SLEDAI-2K, cHDAS, and prednisolone dose, provides a more accurate representation of real-world disease burden than cross-sectional analyses. Furthermore, the systematic collection of clinical, laboratory, and treatment data at 3–6 month intervals and at each visit during active disease or flares enabled the development of a comprehensive longitudinal dataset. These features strengthen the reliability of our evaluation of clinical outcomes, including organ damage and treatment effects. Conclusion Our study highlights the importance of close monitoring and optimal disease management in patients with cSLE who experience HDAS. Prolonged exposure to high disease activity was strongly associated with both an increased risk of, and a shorter time to, damage accrual. Notably, patients with NPSLE were identified as having a significantly higher risk of damage accrual. These findings underscore the critical need for early, sustained, and effective disease control to minimize long-term complications and improve overall patient outcomes.