Validity of protein-osmolality versus protein-creatinine ratios in the estimation of quantitative proteinuria from random samples of urine in children.

Background: Proteinuria is an important marker of kidney disease. Simple methods to determine the presence of proteinuria in a semiquantitative fashion require measurement of either a protein-creatinine or protein-osmolality ratio.

Methods: Urine samples from 134 healthy infants and children and 150 children from the pediatric nephrology practice were analyzed to develop normative data for protein-osmolality ratios on random urine samples and compare protein-osmolality with protein-creatinine ratio as a predictor of 24-hour urine protein excretion. Children were grouped according to age. Three groups were established: infants (<2 years), younger children (2 to 8 years), and older children (9 to 18 years). An adult cohort was similarly analyzed.

Results: For healthy children older than 2 years, the optimal value discriminating normal from abnormal protein excretion was determined to be a protein-osmolality ratio of 0.15 mg x kg H2O/mOsm. L; for children between 2 and 8 years old, 0.14; and for children older than 8 years, 0.17 (P = not significant between age groups). The corresponding optimal cutoff value for protein-creatinine ratio for the entire group of children older than 2 years is 0.20. Area under the curve analysis of receiver operator characteristic curves showed protein-creatinine ratio was superior to protein-osmolality ratio for predicting abnormal amounts of proteinuria in children and adolescents (P < 0.0001). In adults, both ratios are equally accurate.

Conclusion: Given the superiority of protein-creatinine ratio in children, it would be appropriate to screen urine samples for proteinuria using protein-creatinine ratio rather than protein-osmolality ratio.