The use of urinary red cell morphology to determine the source of hematuria in children.

Urinary red cell morphology has been used to indicate the source of renal tract bleeding. A double blind study was undertaken to evaluate the reliability of urinary red cell morphology in predicting the source of hematuria in a pediatric population. Two independent observers used phase-contrast microscopy, one also using Nomarski differential interference microscopy, to examine 101 urine specimens from 90 pediatric patients with hematuria. 28 cases were subsequently excluded because the clinical diagnosis was uncertain. In the remaining 62 patients the laboratory diagnosis based on the erythrocyte morphology was compared with the primary clinical diagnosis. Observer 1 documented dysmorphic red cells in 40 of 42 patients with glomerulonephritis, and isomorphic red cells in 19 of the 20 cases of non-glomerular bleeding. Observer 2 documented dysmorphic red cells in 39 of the 42 glomerular lesions, and isomorphic red cells in all of the non-glomerular lesions. This represents a sensitivity for predicting glomerular hematuria of 95% for observer 1 and 93% for observer 2, and a specificity of 95% and 100% respectively. There was no definite advantage in the use of Nomarski compared to phase-contrast microscopy. These results confirm the accuracy of this technique in predicting the source of hematuria in pediatric patients.