The Captopril Test and Renovascular Hypertension

Ideal screening tests for disease should be highly sensitive and specific, safe, cheap, and simple. The captopril test, as described by Muller et al in 1986¹in a retrospective analysis of the Cornell University (New York, NY) experience, seemed to have most of these attributes and offered the hope of resolution of a particularly knotty clinical problem, the identification of renovascular disease in hypertensive patients. In this issue of theArchives, however, two reports of prospective evaluations of the captopril test in hypertensives reach essentially opposite conclusions regarding its utility: Frederickson et al²recommend it, while Postma et al³do not. That two well-qualified investigative groups should be in such marked disagreement should sound a cautionary note to internists seeking to use the captopril test in the diagnosis of renovascular hypertension, and, more broadly, the dissonance should emphasize the pitfalls inherent in transferring clinical research to clinical