Reliability of different expert systems for profiling proteinuria in children with kidney diseases.

This study was designed to compare three urinary protein expert systems for profiling proteinuria in children with kidney diseases. Freshly voided urine specimens were collected from 61 children with glomerular diseases, 19 children with tubular diseases and 25 healthy children aged 3-16 years. The urinary protein expert systems were: (1) albumin/total protein ratio (APR), (2) alpha-1-microglobulin/alpha-1-microglobulin + albumin algorithm (AAA), and (3) the complex urine protein expert system (UPES, PROTIS) algorithm. APR correctly identified glomerular proteinuria in 47/61 children, tubular proteinuria in 16/19 children and normal proteinuria in 23/25 healthy children. AAA correctly identified glomerular proteinuria in 61/61 children and tubular proteinuria in 18/19 children, and 25/25 healthy children were characterized as having no abnormal proteinuria. AAA was not influenced by the stage of chronic kidney disease. UPES differentiated the type of proteinuria in children with glomerular diseases into glomerular (50/61 patients) and mixed glomerulo-tubular (6/61 patients). Tubular proteinuria was identified in 16/19 patients and described as mixed glomerulo-tubular proteinuria in 3/19 patients. Mixed glomerulo-tubular proteinuria was found only in children with chronic kidney disease stages 2-5 of glomerular and tubular diseases. In conclusion, the AAA and UPES had the highest accuracy levels.