Pitfalls in the diagnosis of phaeochromocytomaBMJ 2004; 328 doi: https://doi.org/10.1136/bmj.328.7440.629 (Published 11 March 2004) Cite this as: BMJ 2004;328:629
- L Zendron, fellow1,
- J Fehrenbach, resident1,
- C Taverna, fellow, oncology2,
- M Krause (), attending physician1
- 1 Department of Internal Medicine, Kantonsspital Münsterlingen, CH-8596 Münsterlingen, Switzerland
- 2 Division of Oncology, Department of Internal Medicine, University Hospital, Rümistrasse 100, CH-8091 Zürich, Switzerland
- Correspondence to: M Krause
Phaeochromocytoma, a rare endocrine disorder causing arterial hypertension, is found in less than 0.1% of patients with hypertension.1 The diagnosis is excluded by the repeated finding of normal catecholamine (for example, dopamine) concentrations in urine or blood. If catecholamine concentrations are increased in combination with an adrenal mass, phaeochromocytoma is usually confirmed by an increased uptake of a radiolabelled catecholamine precursor in the region of the adrenal tumour. We present a case in which the clinical presentation and the results of laboratory, radiological, and nuclear investigations led to an incorrect diagnosis of phaeochromocytoma.
A 73 year old woman presented with severe headache, tinnitus, and a blood pressure of 240/140 mm Hg. She had been treated for arterial hypertension and diabetes mellitus for over 10 years. During a search for secondary hypertension, abdominal ultrasonography showed an enlargement of the right adrenal gland of 3.5 cm. The dopamine level in a 24 hour urine was 25 252 nmol (upper normal limit 2500 nmol). This was confirmed by a second analysis (35 628 nmol). Adrenaline and noradrenaline concentrations were within normal ranges. Magnetic resonance imaging confirmed a 2.5 × 3.5 × 4 cm mass in the right adrenal gland, and a scan using metaiodobenzylguanidine labelled with iodine-123 showed isolated uptake of radionuclide in the same region (figure). A dopamine secreting, possibly malignant, right sided phaeochromocytoma was postulated, and the patient underwent right adrenalectomy. The procedure was uneventful. Histology, however, showed a large adenoma of the adrenal cortex; a phaeochromocytoma could not be detected. Postoperatively, the patient still required antihypertensive treatment. The urinary excretion of dopamine remained high at 43 068 nmol daily, and a repeat 123I-metaiodobenzylguanidine scan showed the same hot spot.
The patient was taking 125 mg levodopa with carbidopa twice a day for restless legs. This was stopped. A 24 hour urine sample three weeks later contained 4019 nmol dopamine, but this reverted to normal values four months later. Re-examination of the 123I-metaiodobenzylguanidine scan and magnetic resonance imaging showed that the hot spot was located well below the adrenal mass and that it was caused by delayed excretion of the radionuclide due to a benign dilation of the right renal pelvis. The patient's blood pressure is now satisfactorily controlled with four antihypertensive drugs.
This patient presented with poorly controlled blood pressure, which prompted the clinicians to search for secondary hypertension. Three classic pitfalls, however, led to the misdiagnosis of phaeochromocytoma. The first pitfall was that the finding of an adrenal tumour was related to hypertension. Adrenal tumours are incidentally found in 1.5% of computed tomograms; in a large survey, only 4.2% were found to be phaeochromocytomas. Thus the presence of an adrenal tumour together with hypertension is a coincidence in most of the cases.2 The second pitfall was the assumption that the high urinary dopamine levels were related to the tumour, and other causes for false positive catecholamine concentrations were not considered.3 4 Our patient was taking a moderate dose of levodopa with carbidopa. After the drug was stopped, dopamine levels returned to normal over a period of several weeks, indicating that this drug was responsible for the increased levels. The specificity of urinary catecholamines in sporadic cases of phaeochromocytoma is only 75%. Recently, the determination of levels of plasma free metanephrines has been recommended, which have a higher specificity and sensitivity for detection of phaeochromocytoma.5 The third pitfall was that a positive 123I-metaiodobenzylguanidine scan was thought to confirm the diagnosis of right sided phaeochromocytoma. The specificity of this scan is more than 95%, although accumulation of radionuclide in a dilated renal pelvis may give a false positive result.6 7
In this case, the sequence of positive tests misled the clinicians into suspecting a phaeochromocytoma. In a population of patients with therapy resistant hypertension and an adrenal mass, the prevalence of phaeochromocytoma is around 30%.5 The finding of increased levels of urinary catecholamines (sensitivity 91%, specificity 75%) double this probability.5 If 60% are taken as a pretest probability for next test, a positive 123I-metaiodobenzylguanidine scan (sensitivity 100%, specificity of 95%) increases the probability to 97%.8 A correct diagnosis of phaeochromocytoma depends on a critical interpretation of each diagnostic test, a thorough review of anatomical and functional examinations, and careful history taking, including drugs.
Careful drug history and critical review of all tests are essential to diagnose phaeochromocytoma
Contributors LZ wrote the first draft of the paper; CT, JF, and MK contributed to revisions. LZ, JF, and MK cared for the patient while in hospital. LZ will act as guarantor for the paper.
Competing interests None declared.