Early proteinuria response: a valid real-life situation predictor of long-term lupus renal outcome in an ethnically diverse group with severe biopsy-proven nephritis?

Abstract

Objective

Two recent important lupus nephritis trials reported that proteinuria was a good predictor of renal outcome in Caucasians, but data on real-life situation, other races and severe nephritis are lacking to substantiate this finding as a simple test to guide clinical practice. The aim of this study was to validate proteinuria as a predictor of long-term renal outcome in real-life situation in a racially diverse group of patients with severe nephritis.

Methods

Proteinuria, serum creatinine (SCr) and urine red blood cells were assessed at baseline and after 3, 6 and 12 months, as early predictors of long-term renal outcome (SCr <1.5 mg/dL at 7 years), in 94 patients with biopsy-proven lupus nephritis. The parameter performance and cut-off values were computed by receiver operating characteristic curves. Kaplan-Meier curves were used to validate the parameter.

Results

A proteinuria <0.8 g/24 hours at 12 months was the best single predictor of long-term renal outcome (sensitivity 90%, specificity 78%, positive predictive value 67%, negative predictive value (NPV) 94% and area under the curve 0.86; p<0.001). Addition of other variables to proteinuria such as SCr and haematuria at 12 months did not improve its performance. The proteinuria cut-off value of 0.8 g/24 hours at 12 months was a good predictor of 7-year renal survival (years free of dialysis) for patients with pure membranous (p=0.005) and proliferative nephritis (p=0.043), as well as black (p=0.002) and white race (p=0.001), anti-dsDNA positive (p=0.001) and anti-dsDNA negative (p=0.04) and male (p=0.028) and female (p=0.003) patients.

Conclusion

We provided novel evidence that, in a real-life situation, proteinuria at 12 months of follow-up was the single best predictor of renal outcome at 7 years for an ethnically diverse group of patients with severe nephritis and a valid parameter for distinct histological classes, races, genders and anti-dsDNA profiles. The remarkably high NPV obtained reinforces its recommendation as the ideal predictor for clinical practice, since it is of low cost, easy to interpret, non-invasive and widely available.