Abstract
BACKGROUND: Lupus nephritis (LN) is a crucial organ involvement in systemic lupus erythematosus (SLE). Patients with LN have higher morbidity and mortality rates than those without. Among all patients with LN, 20-40% had delayed onset, but the data for patients with juvenile-onset SLE (jSLE), who have a higher percentage of LN than patients with adult-onset SLE (aSLE), were limited. This study aimed to determine the risk factors for subsequent LN in patients with jSLE. METHODS: A retrospective cohort study was conducted between 2008 and 2018 in a single tertiary medical centre. Patients with diagnosed jSLE were reviewed. We investigated those without LN at diagnosis and whether they developed LN afterward. The primary outcome was the development of subsequent LN. Clinical manifestations at diagnosis, serial laboratory data, and treatments were reviewed during follow-up periods. RESULTS: Among the 48 patients with jSLE without initial LN, 20 developed subsequent LN later (Group 1), whereas 28 remained free of LN (Group 2). There was no difference in the percentage of initial manifestations except for more discoid rashes in Group 2 patients. In the Cox regression model, elevated average anti-double-stranded DNA (dsDNA) antibody, low average serum complements, and high average erythrocyte sedimentation rate (ESR) levels during follow-up were predictors of subsequent LN. After adjusting for these factors in multivariable analyses, only high average anti-dsDNA antibody and high average ESR levels remained predictive of subsequent LN. For every 100 IU/ml increase in anti-dsDNA antibody, the risk for subsequent LN in jSLE increases by 1.29 times (hazard ratio = 1.29, 95% confidence interval 1.055-1.573). CONCLUSION: Persistently high anti-dsDNA antibody and ESR levels during the follow-up period were risk factors for subsequent LN in patients with jSLE.