Objectives: The aims of this study were to determine the performance of each variable, to define the optimal diagnostic thresholds and to determine the relative value of assaying chromogranin A (CgA).
Design: Prospective study.
Methods: Two groups of patients were studied: a control group of 71 patients and a group of 63 patients with a histologically-proven pheochromocytoma (52 pheochromocytomas and 14 paragangliomas). Fourteen of the patients had a family history of the disease. Eleven variables were assayed in each patient, i.e. the plasma and urinary concentrations of amines and their derivatives, and the CgA serum concentration.
Results: The study of the control group showed that all the serum assays gave false positive results (from 6 to 23%), as did four of the six urinary assays (from 2.9 to 12.3%). The areas under the receiver operating characteristic curves varied from 0.689 to 0.992. The variables relating to the epinephrine pathway were significantly less expressed in the hereditary diseases than in the sporadic cases. The diagnostic thresholds of the three most efficient variables have been raised.
Conclusions: Plasma determinations of metanephrines are now an easy and convenient tool for the diagnosis of pheochromocytoma. However, in our study the best specificity was obtained with the urinary tests rather than with the plasma assays while the highest sensitivities were for the normetanephrine assays. The assay of CgA was highly efficient in diagnosing pheochromocytomas in the absence of renal insufficiency. By combining it with fractionated metanephrine assays, the sensitivities of the latter were increased.