Background/Purpose: End-stage renal disease (ESRD) develops in 10-30% of patients with lupus nephritis (LN) despite standard treatment. A biomarker with excellent sensitivity and specificity to predict the evolution of LN would determine the intensity and duration of treatment as a primary end point in clinical trials. It was shown that proteinuria at 12 months has a sensitivity of 71% and specificity of 75% to predict renal outcomes at 7 years. There is evidence that serum albumin could predict renal evolution, the basis of this premise is the clinical observation that, despite proteinuria, preserved serum albumin is associated with sustained renal function. The objective of this study is to evaluate the association between baseline serum albumin and at 12 months with renal evolution at two years.

Methods: Cross-sectional, retrospective study in patients with SLE (SLICC 2012) with LN staged by biopsy, between January/2015 to January/2019. The levels of serum albumin, complement, urinary sediment and proteinuria were evaluated at diagnosis and at 12 months. An adverse renal outcome (ARO) was defined as the doubling of baseline serum creatinine, or creatinine >4 mg/dL if the initial one was >2.5 mg/dL, ESRD requiring hemodialysis (HD) therapy or transplantation. ROC curve analysis was performed for albumin value below 3 g/dL as cut value. Patients had to have at least one additional follow-up visit 24 months after diagnosis.

Results: We included 62 patients, 87% women with a mean age of 32 years (SD +/- 11). FAN positive in all patients. Anti-DNA 63%, Anti-Sm 24%, Anti-Ro 37%. ACL IgM 5%, IgG 6.7%. B2-IgM glycoprotein 5%, IgG 8.3%. LA 10%. Hypocomplementemia 89% (C3 63%, C4 89%). LN: II 16%; III 26%; IV 45%; V 11%; VI 2%. 54% of patients with LN received pulses of steroids. Induction therapy: CFD 61%, RTX 6.5% and MMF 25%. Maintenance therapy: MMF 70%, AZA 20%. They received HCQ 96.7%.
At the diagnosis of LN: proteinuria 4.3 g/day (SD +/- 3), serum albumin 3.4 g/dL (SD +/- 0.9 CI 95% 2.6-4.2); serum albumin at 12 months 4.2 g/dL (SD +/- 0.57 IC 95% 3.7-4.6). Basal hypoalbuminemia association with ARO p = 0.44 (creatinine increase: p = 0.45, ESRD p = 0.43, HD p = 0.93). Hypoalbuminemia at 12 months with ARO p = 0.003 (Creatinine increase p = 0.007, ESRD p = 0.001, HD p = 0.61). A ROC curve is made taking albumin values less than 3 g/dL as a cut-off value AUC = 0.77 p = 0.001; sensitivity 61% and specificity 93%.

Conclusion: We found an association between the level of serum albumin at 12 months and renal evolution at two years, so it should be taken into account as it is a quick, easy and cheap study. Hypoalbuminemia has a moderate sensitivity and good specificity to predict poor renal evolution. The association between serum albumin at diagnosis and renal evolution was not significant. Prospective studies in large cohorts are necessary to validate this association, investigate causality and its usefulness in monitoring and therapeutic decisions.

« Back to 2019 ACR/ARP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/serum-albumin-as-a-long-term-predictor-of-renal-evolution-in-lupus-nephritis/