Background: The purpose of this study was to evaluate the correlation and consistency between urine protein/creatinine ratio (UPCR) and 24-h urine protein (24HUPr) in children, and to determine cutoff values of UPCR relative to 24HUPr at 100 mg/m2/d (≥ 100 mg/m2/d as pathological proteinuria) and 1000 mg/m2/d (≥ 1000 mg/m2/d as nephrotic-range proteinuria).
Methods: Three hundred sixty-six children were enrolled, including 81 controls, 109 with Henoch-Schönlein purpura nephritis, 167 with nephrotic syndrome, 5 with IgA nephropathy, and 4 with lupus nephritis. Patients were divided into three groups: normal group; non-nephrotic-range proteinuria group; nephrotic-range proteinuria group. The cutoff values of UPCR in predicting the different levels of proteinuria were determined using ROC curve analysis.
Results: UPCR was positively correlated with 24HUPr (r = 0.915, p < 0.01). Bland-Altman diagrams showed that UPCR and 24HUPr had good consistency, and > 95% spots of UPCR and 24HUPr were within 95% confidence intervals. Relative to 24HUPr at 100 mg/m2/d, the cutoff value of UPCR (0.18 g/g Cr) had the highest sensitivity (94%) and specificity (98.8%) which is close to 0.2 g/g Cr as proposed by the American College of Rheumatology. Relative to 24HUPr at 1000 mg/m2/d, the cutoff value of UPCR (2.09 g/g Cr) had the highest sensitivity (92.1%) and specificity (92.1%) which is close to the 2.0 g/g Cr proposed in KDIGO guidelines.
Conclusions: UPCR showed strong correlation and consistency with 24HUPr for evaluating levels of proteinuria in children. UPCR < 0.2 g/g Cr can be considered a criterion for normal-range proteinuria. Instead of 24HUPr ≥ 1000 mg/m2/d, UPCR ≥ 2.0 g/g Cr can be considered a criterion for nephrotic-range proteinuria or nephrotic syndrome in children.
Keywords: 24-h proteinuria quantity; Consistency; Correlation; Cutoff value; Urine protein/creatinine ratio.