Traditional methods of noninvasively evaluating patients for renal injury do not accomplish the following tasks: reliably distinguish potentially treatable forms of acute renal failure from acute tubular necrosis; provide a sensitive indicator of early allograft rejection in renal transplant recipients, particularly those in the pediatric age group; provide an early warning of incipient drug-induced nephrotoxicity; or serve as an adequate screening test for renal injury due to exposure to occupational or environmental toxins, especially heavy metals. Because of this, considerable effort has been devoted to the development of assays to satisfy these needs. Three approaches include measurement in the urine of low-molecular-weight plasma proteins such as beta 2-microglobulin; a variety of kidney-derived enzymes, such as L-alanine aminopeptidase and N-acetyl-beta-D-glucosaminidase; and specific renal antigens using immunologic detection. The first two of these have not proved to be adequately sensitive or specific, complicated by the frequent loss of activity associated with the physicochemical characteristics of the urine or the presence of pyuria. Despite this, useful information has been obtained. In particular, assays of beta 2-microglobulin urinary excretion and retinol binding protein appear to have clinical utility that should be pursued. Recent experience with a monoclonal antibody-based assay for a unique proximal tubular antigen, the adenosine deaminase binding protein, suggests that a battery of such assays, each directed against an antigen localized to a particular segment of the nephron, may be particularly useful.