Role of venous sampling in locating a phaeochromocytoma.

Selective venous sampling was performed in 31 patients in whom the diagnosis of phaeochromocytoma was suspected on clinical and biochemical grounds. Data from samples assayed for their adrenaline and noradrenaline content using a radioenzymatic technique were used to confirm or refute the suspected diagnosis. In 19 of the 31 patients a phaeochromocytoma was subsequently removed surgically, and the remaining 12 patients are now thought not to have tumours (mean follow up period: four years). Analysis of the assay data shows that selective venous sampling correctly identified the presence of a tumour in all 19 patients, and correctly excluded the diagnosis in 11 of the 12 remaining patients--an overall success rate of 97%. Success rates of 88% for arteriography and 84% for computed tomography were recorded, though these investigations were not performed in all patients. Ultrasound and intravenous urography were much less accurate. On the basis of this study a sequence of investigation is proposed for patients with a suspected phaeochromocytoma. Computed tomography occupies a central role in this sequence with venous sampling (and occasionally other techniques) being used only as complementary investigations when specific indications for their use exist.