Childhood-Onset Lupus Nephritis in the Childhood Arthritis and Rheumatology Research Alliance Registry: Short-Term Kidney Status and Variation in Care

Abstract

Objective: The goal was to characterize short-term kidney status and describe variation in early care utilization in a multicenter cohort of patients with childhood-onset systemic lupus erythematosus (cSLE) and nephritis.

Methods: We analyzed previously collected prospective data from North American patients with cSLE with kidney biopsy-proven nephritis enrolled in the Childhood Arthritis and Rheumatology Research Alliance (CARRA) Registry from March 2017 through December 2019. We determined the proportion of patients with abnormal kidney status at the most recent registry visit and applied generalized linear mixed models to identify associated factors. We also calculated frequency of medication use, both during induction and ever recorded.

Results: We identified 222 patients with kidney biopsy-proven nephritis, with 64% class III/IV nephritis on initial biopsy. At the most recent registry visit at median (interquartile range) of 17 (8-29) months from initial kidney biopsy, 58 of 106 patients (55%) with available data had abnormal kidney status. This finding was associated with male sex (odds ratio [OR] 3.88, 95% confidence interval [95% CI] 1.21-12.46) and age at cSLE diagnosis (OR 1.23, 95% CI 1.01-1.49). Patients with class IV nephritis were more likely than class III to receive cyclophosphamide and rituximab during induction. There was substantial variation in mycophenolate, cyclophosphamide, and rituximab ever use patterns across rheumatology centers.

Conclusion: In this cohort with predominately class III/IV nephritis, male sex and older age at cSLE diagnosis were associated with abnormal short-term kidney status. We also observed substantial variation in contemporary medication use for pediatric lupus nephritis between pediatric rheumatology centers. Additional studies are needed to better understand the impact of this variation on long-term kidney outcomes.

Figure 1.

Recorded medication ever use across the total cohort of biopsy-proven lupus nephritis patients in the Childhood Arthritis and Rheumatology Research Alliance (CARRA) Registry (n = 222). ACE-I = angiotensin-converting enzyme inhibitor; ARB = angiotensin-receptor blocker; IV = intravenous.

Figure 2.

Center-level variation in ever use of mycophenolate only, cyclophosphamide, rituximab, cyclophosphamide-rituximab combination, or other disease-modifying antirheumatic drugs (DMARDs) for patients with biopsy-proven proliferative nephritis. Includes 15 pediatric rheumatology centers, Sites A–O, with ≥5 patients enrolled in the Childhood Arthritis and Rheumatology Research Alliance (CARRA) Registry with proliferative nephritis. A hierarchical categorization schema was applied so that patients categorized as cyclophosphamide and/or rituximab may have also ever received mycophenolate or other DMARDs but not vice versa. Patients with both mycophenolate and other DMARD ever use recorded were categorized as mycophenolate. Patients included in the other DMARD category had recorded ever use of azathioprine, methotrexate, or hydroxychloroquine only.