Interobserver reproducibility in the interpretation of captopril renograms from patients suspected of having renovascular hypertension.

This work was undertaken to identify, in hypertensive patients suspected of having renovascular hypertension (RVH), scintigraphic images and curves whose interpretation could be difficult. The baseline and postcaptopril Tc-99m MAG 3 images and curves from 10 patients clinically suspected of having RVH were retrospectively selected to have a wide sample of images and curves. The data and a multiple-choice questionnaire were presented as a PowerPoint file, which was sent electronically to Belgian nuclear medicine physicians; they had to answer if the images and curves were or were not suggestive of the diagnosis of RVH or if it was impossible to answer. There was near-complete interobserver agreement for normal baseline and postcaptopril data, for unilateral postcaptopril delayed transit and/or impaired uptake. Agreement was not as good for bilateral and symmetrical changes, for mild changes, or in cases of impaired function. This study shows that the analysis by the Internet of interobserver agreement is an easy and rapid method allowing one to identify difficult to interpret pre- and postcaptopril renographic data.