Spot urine protein/creatinine ratios are unreliable estimates of 24 h proteinuria in most systemic lupus erythematosus nephritis flares

Abstract

The diagnosis of glomerulonephritis flares in systemic lupus erythematosus (SLE) is usually based on whether the magnitude of proteinuria has changed. Our study tests two methods to assess proteinuric change: protein/creatinine (P/C) ratios of intended 24-h urine collections or that of spot urine samples. Sixty-four patients with glomerulonephritis due to SLE followed in the Ohio SLE Study provided bimonthly paired spot and intended 24-h urine collections. Completeness of each collection was estimated as the ratio of the measured creatinine to the expected creatinine based upon Cockroft-Gault. Intended 24-h urine collections with measured/expected creatinine ratios between 0.5 and 0.9 (237 samples overall) showed ratios that were not significantly different from ratios of complete 24-h urine collections with ratios of 0.9-1.1 (159 samples). To compare spot and 24 h P/C ratios, we randomly selected pairs of samples with measured/expected ratios above 0.75. Consistent with previous studies, spot and 24-h urine P/C ratios showed good correlation over the range of values as well as reasonably strong concordance. Over the range of most SLE glomerulonephritis flares, however, correlation was present but concordance was poor. Our work suggests that the use of spot urine P/C ratios will yield more false-positive and -negative diagnoses of glomerulonephritis flares in patients with SLE than the ratio in 24-h urines.