Item-specific factors associated with damage accrual in systemic lupus erythematosus: insights from a multiethnic, multinational Latin American cohort

Abstract

Objective: To identify factors associated with individual Systemic Lupus International Collaborating Clinics/American College of Rheumatology Damage Index (SDI) accrual items in 1385 patients from the Latin American SLE incident cohort (GLADEL (Grupo Latino Americano Del Estudio de Lupus)).

Methods: Longitudinal cohort study. SDI assessed yearly, and damage accrual defined as a SDI change in any individual item during follow-up. Sociodemographic and clinical characteristics at entry, along with regular medication use up to the month prior to each damage item accrued, were analysed using multivariate Cox proportional hazard models with time-varying hazard risk effect performed. Models for overall damage and 10 items with at least 28 patients per item are reported.

Results: New damage was accrued by 658 patients with SLE (48%) over a median of 54 months (p25-75: 29-71) of follow-up. Mestizo ethnicity, recent SLE Disease Activity Index (SLEDAI) and immunosuppressants were predictors of overall damage accrual, while antimalarials and rural residence were protectors; mixed effects were observed for SDI >0 and glucocorticoids (GCs) use. Protectors for scarring chronic alopecia (23.7% at 7 years follow-up) included older age, longer disease duration, diagnostic delay, SDI >0 and cytotoxic use. Proteinuria >3.5 g/24 hours (11.5% at 7 years follow-up) was associated with protective factors like older age, longer disease duration and higher GCs dose, while risk factors were Mestizo ethnicity, low medical coverage, higher SLEDAI scores and cytotoxic use. In addition, Mestizo ethnicity was a risk factor for estimated glomerular filtration rate <50% and end-stage renal disease (ESRD) whereas antimalarial protected from ESRD. GCs were risk factors for pericarditis, retinal change or optic atrophy but provided protection against proteinuria >3.5 g/24 hours. Recent disease activity and cytotoxic use were significant risk factors for additional items.

Conclusions: Factors associated with damage accrual differ substantially by individual SDI items. Mestizo ethnicity and recent disease activity increased renal risks, while antimalarials were protective. GCs showed mixed effects. Item-specific strategies are crucial to mitigate long-term damage.

Keywords: Outcome Assessment, Health Care; Risk Factors; Systemic Lupus Erythematosus.

Figure 1. Survival curves for damage accrual. (a) Kaplan-Meier curve and (b) Kaplan-Meier curves. Log-rank test p value <0.001. SDI, Systemic Lupus International Collaborating Clinics/American College of Rheumatology Damage Index.

Figure 2. Hazard ratios from the univariate Cox regression models. Accrual damage (de novo) in the GLADEL cohort. ~Time variant hazards risk. *HR Ethnic Group Mestizo versus white. Exception HR for seizures and retinal changes is Others versus white Latin American. **HR socioeconomic status lower/middle lower versus upper. Exception: retinal change or atrophy that SES is lower/middle lower versus middle plus upper/middle upper. ***SLEDAI (SLE Disease Activity Index) HR for each additional point score. #SDI for each additional point score. ¹HR for an additional 10 mg of prednisone equivalent dose in the prior month. ²HR for an additional 100 mg of cumulative prednisone equivalent dose until the prior month. ³Antimalarials, cytotoxic and other immunosuppressant use until the prior month. ACR, American College of Rheumatology; GLADEL, Grupo Latino Americano Del Estudio de Lupus; SES, socioeconomic status; SLEDAI, SLE Disease Activity Index; SLICC, Systemic Lupus International Collaborating Clinics.

Figure 3. Hazard ratios from the Cox regression models with time varying effects. Association of factors on accrual damage by item during follow-up. Predictors for damage accrual items are depicted in colours: significant strong protective factors with an HR <0.5 in bright green and lighter for moderate or low protection, and strong risk predictors with an HR >2 in bright red, offering lower protection in shades of red. Boxes with divisions indicate a time-varying effect presence. *HR Ethnic Group Mestizo versus white. Exception is HR for seizures requiring therapy for 6 months and retinal change; the risk is for Others versus white. **HR socioeconomic status lower/middle lower versus upper. Exception HR for SES lower/middle lower versus middle and middle/upper is retinal change items. ***SLEDAI HR for each additional one points score. #SDI, HR for each additional one points score. ¹HR for each additional 10 mg of prednisone dose in the prior month to the accrual damage. ²HR for each additional 100 mg of cumulative prednisone dose until the prior month to the accrual damage. ³Antimalarials (chloroquine or hydroxychloroquine), cytotoxic (cyclophosphamide) and other immunosuppressants (azathioprine or methotrexate), some patients received mycophenolate mofetil until the prior month to damage accrual. ACR, American College of Rheumatology; SES, socioeconomic status; SLEDAI, SLE Disease Activity Index; SLICC, Systemic Lupus International Collaborating Clinics.