A comparison of plasma-free metanephrines with plasma catecholamines in the investigation of suspected pheochromocytoma

Abstract

To compare the diagnostic performance of plasma metanephrines by ELISA and plasma catecholamine measurements by HPLC in patients selected for clonidine suppression testing. Plasma catecholamines adrenaline (ADR) and noradrenaline (NOR) were measured by HPLC and metanephrine with normetanephrine (NMN) by ELISA (n = 67). The diagnostic performance of metanephrines was determined by receiver operating characteristic (ROC) curve analysis. Phaeochromocytoma was confirmed by histological analysis in 14 patients and excluded in 53 patients by a negative clonidine suppression test (CST), abdominal computerized tomography scan and clinical follow-up (median 2.5 years). A sensitivity and specificity of 100 and 96%, respectively, was obtained by using our current CST diagnostic criteria for ADR and NOR values. ROC curve analysis revealed optimum sensitivity and specificity for plasma-free metanephrines using a threshold of 784 pmol/l at baseline and 663 pmol/l at 180 min. Baseline measurements of metanephrine with NMN showed 100% sensitivity and 98% specificity, as assessed by ROC curve analysis-derived criteria or when evaluated against published decision thresholds. A sensitivity and specificity of 100% was obtained for the combined measurements of metanephrine with NMN at 180 min. Plasma metanephrines (metanephrine with NMN) were equally effective as plasma catecholamines during CST. This study supports the use of measuring plasma metanephrines by ELISA as a less labour-intensive and equally effective biochemical test for phaeochromocytoma in patients with a high clinical suspicion. There was still overlap between groups with and without phaeochromocytoma at baseline under controlled conditions and clinically some patients still need to undergo clonidine suppression testing.