A blind controlled trial of phase-contrast microscopy by two observers for evaluating the source of haematuria.

Two observers, unaware of each other's findings and of the diagnosis, examined red cells in the urine of 109 patients by phase-contrast microscopy. All specimens were examined uncentrifuged by inverted microscopy, and 48 of the 109 were also examined under a light microscope after centrifugation. We were unable to confirm with either method the close correlation between red-cell morphology and diagnosis reported in previous studies, and our 2 observers differed in their interpretation on 38% of occasions. Proteinuria associated with the presence of casts in uncentrifuged urine examined by inverted microscopy was a better indication than red-cell morphology of renal parenchymal disease.