We used a two-round consensus panel method to derive and rate the appropriateness of comprehensive sets of detailed clinical indications for performing carotid endarterectomy. Before meeting, nine nationally influential physicians rated 675 indications; after review and discussion, they rated 864. The method did not force unanimity; our purposes were not only to encourage agreement but also to uncover areas of disagreement concerning the procedure's appropriate use. The panelists agreed on the level of appropriateness for 54 per cent of the final 864 indications and disagreed on 18 per cent. Ratings were reliably reproduced six to eight months after the completion of the process. The physicians' indications and ratings were consistent with those in the literature, and statistical analysis demonstrated that they followed logical clinical rationale. We conclude that consensus methods that do not force agreement can be used with panels of physicians to produce detailed, reliable, and valid indications. They can also identify medically controversial reasons for using a procedure that can serve as a starting point for a research agenda.