Biochemical diagnosis of chromaffin cell tumors in patients at high and low risk of disease: plasma versus urinary free or deconjugated o-methylated catecholamine metabolites

Abstract

BACKGROUND: Measurements of plasma or urinary metanephrines are recommended for diagnosis of pheochromocytoma and paraganglioma (PPGL). What test offers optimal diagnostic accuracy for patients at high and low risk of disease, whether urinary free metanephrines offer advantages over deconjugated metanephrines, and what advantages are offered by including methoxytyramine in panels all remain unclear. METHODS: A population of 2056 patients with suspected PPGLs underwent prospective screening for disease using mass spectrometric-based measurements of plasma free, urinary deconjugated, and urinary free metanephrines and methoxytyramine. PPGLs were confirmed in 236 patients and were excluded in others on follow-up evaluation. RESULTS: Measurements of plasma free metabolites offered higher (P < 0.01) diagnostic sensitivity (97.9%) than urinary free (93.4%) and deconjugated (92.9%) metabolites at identical specificities for plasma and urinary free metabolites (94.2%) but at a lower (P< 0.005) specificity for deconjugated metabolites (92.1%). The addition of methoxytyramine offered little value for urinary panels but provided higher (P < 0.005) diagnostic performance for plasma measurements than either urinary panel according to areas under ROC curves (0.991 vs 0.972 and 0.964). Diagnostic performance of urinary and plasma tests was similar for patients at low risk of disease, whereas plasma measurements were superior to both urinary panels for high-risk patients. CONCLUSIONS: Diagnosis of PPGLs using plasma or urinary free metabolites provides advantages of fewer falsepositive results compared with commonly measured deconjugated metabolites. The plasma panel offers better diagnostic performance than either urinary panel for patients at high risk of disease and, with appropriate preanalytics, provides the test of choice. Measurements of methoxytyramine in urine show limited diagnostic utility compared with plasma.