An apparent phaeochromocytoma and abnormal thyroid function tests

Abstract

Abstract

A 41-year-old lady presented with classical symptoms of a phaeochromocytoma and markedly elevated urinary adrenaline concentration. Urinary fractionated metanephrine and normetanephrine concentrations were within the normal reference range. Computed tomography of the abdomen/pelvis and I¹²³ meta-iodobenzylguanidine scans were negative. Factitious adrenaline use was established as the diagnosis.

Introduction

Phaeochromocytomas are catecholamine-secreting paragangliomas that arise from chromaffin cells of the adrenal medulla and the sympathetic ganglia. They are rare tumours, probably occurring in <0.2% of patients with hypertension. ¹ Clinically, phaeochromocytoma is usually suggested by the history in a symptomatic patient, discovery of an incidental adrenal mass or the family history in a patient with familial disease.

Case report

A 41-year-old housewife presented with palpitations, sweating, facial flushing and headaches. Her background history included hypertension, urticarial vasculitis, recurrent urinary tract infections and primary hypothyroidism. Her medications included doxazosin, metoprolol, cimetidine, amitriptyline, gabapentin, levocetirizine and tramadol. She had a history of acute angioedema and had in the past been prescribed Anapen^® (Celltech, Bath, UK) by her immunologist for an acute anaphylactic episode.

Thyroid function tests were normal. Three consecutive 24 h urine collections were sent for catecholamines, two of which showed elevated free adrenaline concentrations of 1271 and 415 nmol/24 h (reference range <230 nmol/24 h), respectively.

A renal ultrasound and non-contrast computed tomography (CT) scan of her abdomen and pelvis were normal. Endocrine consult was sought. As beta-blockers and tricyclic antidepressants are known to interfere most frequently with the interpretation of 24 h urinary catecholamines and their metabolites, her beta-blocker and tricyclic antidepressant were stopped for three weeks prior to admission to hospital for three repeat consecutive 24 h urine collections for catecholamines (Table 1, samples 4–6). Throughout her hospital stay she was normotensive with a normal resting heart rate.

Table 1

Urinary free catecholamines and total fractionated metanephrines

Analyte	Sample 1	Sample 2	Sample 3	Sample 4	Sample 5	Sample 6	Reference range
Noradrenaline (nmol/24 h)	204	270	194	155	221	116	<900
Adrenaline (nmol/24 h)	1271	415	44	1635	2775	4896	<230
Dopamine (nmol/24 h)	1478	1501	962	973	1611	1025	<3300
T. Normetadrenaline (nmol/24 h)	2514	2288	1694	973	1593	906	<2800
T. Metanephrine (nmol/24 h)	526	554	376	456	500	328	<1800
HMMA (μmol/24 h)	22.4	23.8	15.1	11.7	20.7	16.8	<45.0
HVA (μmol/24 h)	25.2	34.1	24.1	26.1	35.9	32.2	<40.0
Urine volume (L)	2.8	1.7	1.5	0.9	0.9	1.4

HVA, homovanilic acid; HMMA, 4-hydroxy-3-methoxymandelic acid

Adrenaline concentrations in all three urine collections were markedly elevated, 7–20 times the upper limit of the reference range. Of note, metanephrine, normetanephrine and 4-hydroxy-3-methoxymandelic acid (HMMA) concentrations were within the respective reference ranges. I¹²³ meta-iodobenzylguanidine (MIBG) scintigraphy was normal. In view of the biochemical and radiology results, it was suspected that this woman was surreptitiously adding adrenaline to her urine. The patient was questioned with respect to her use of the Anapen^®. She reported that her angioedema had been well controlled and that she had not received an adrenaline pen for at least two years.

The Anapen^® is an automatic injection device, containing a sterile, pre-loaded syringe of adrenaline. It is designed to deliver a single dose of adrenaline (adult dose is 0.3 mg) in the emergency treatment of severe anaphylactic reactions only. There is 1 mL of adrenaline per device (∼3 adult doses), but for stability reasons, ∼0.7 mL of adrenaline is left in the syringe after use. In the laboratory, a 24 h urine collection from an enthusiastic scientist was analysed for free fractionated catecholamines and metanephrines (unspiked urine). This urine was then spiked with 0.3 mL (300 μg) of adrenaline (spiked urine) and re-analysed for free fractionated catecholamines and metanephrines. The analysis was able to reproduce the patient's biochemical picture (Table 2).

Table 2

Urinary free catecholamines and total fractionated metanephrines in unspiked and spiked 24 h urine collection

Analyte	Unspiked urine (nmol/24 h)	Spiked urine (nmol/24 h)	Reference range (nmol/24 h)
Noradrenaline	77	89	<900
Adrenaline	19	1464	<230
Dopamine	1631	1841	<3300
T. Normetadrenaline	609	636	<2800
T. Metanephrine	494	515	<1800
Urine volume (L)	0.75	0.75

Consultation with the patient's primary care physician and pharmacist revealed that she had in fact received and encashed four prescriptions for Anapens^® in the preceding six months, one just before her admission to hospital for the three urine collections.

The patient was informed of her results and advised that she did not have a phaeochromocytoma. She was followed in the outpatients with three repeat urine collections and on each occasion the urine tests were negative for catecholamines and metanephrines, but still she complained of headaches and palpitations. Examination revealed that she was sweaty and tremulous, blood pressure was 110/70 mmHg, pulse was 80 bpm. She had a palpable small non-tender thyroid gland. Thyroid function tests, confirmed in a second laboratory, revealed a low thyroid-stimulating hormone (TSH) and elevated free thyroxine (fT4) suggestive of thyrotoxicosis (Table 3).

Table 3

Thyroid function test results

Date	fT4 (8.0–21.0 pmol/L)	TSH (0.5–4.2 mIU/L)
3 July 06	32.1	0.1
6 July 06	20.1	0.06
12 July 06	42.1	0.03

fT4, free thyroxine; TSH, thyroid-stimulating hormone

Thyroid peroxidase antibodies were negative. Thyroid Tc^{99 m} isotope scan showed decreased isotope uptake with no evidence of nodular disease. Inflammatory markers were mildly elevated (erythrocyte sedimentation rate 29 mm/h, C-reactive protein 13 mg/L [reference range 1–10]), but were not typical of or consistent with a thyroiditis. The patient said that she had been prescribed l-thyroxine three years previously for mild subclinical primary hypothyroidism (TSH = 5.24 mIU/L, with a normal fT4 of 13.3 pmol/L). The thyroxine therapy had been stopped, she said after 18 months as she had found no clinical benefit from the therapy. Again, contact was made with her primary care physician and pharmacist, both confirmed that she had received and encashed prescriptions for 50 mcg of l-thyroxine in the preceding six months. No treatment was started for her biochemical thyrotoxicosis, but the patient was questioned about the discrepancy between the pharmacist's and her version of events, she had no explanation and suggested that the pharmacy must be wrong. Thyroid function tests, however, normalized spontaneously following this conversation.

Discussion

In patients with phaeochromocytoma, the release of catecholamines by the tumour may be intermittent, or the uptake and degradation of the catecholamines by the tumour may be such that the normal concentrations of catecholamines will be seen in blood and urine, when striking increases in catecholamine metabolites are observed in urine. ² Urinary catecholamines have a 92% sensitivity, whereas urinary fractionated metanephrines have a 97% sensitivity for the diagnosis of sporadic phaeochromocytoma. ³ The measurement of urinary fractionated metanephrines is considered the single most reliable screening test for phaeochromocytoma (almost no false-negative results), specificity is approximately 89% (false-positives are seen in times of severe stress and in patients on monoamine oxidase inhibitors). To maximize diagnostic accuracy, measurement of free catecholamines or total fractionated metanephrines in urine is recommended in the initial biochemical assessment. Plasma free metanephrines have a 99% sensitivity and can be used in the biochemical assessment, and indeed may have proven useful in this case as it would have been much more difficult to interfere with the sample. ³ While our patient was found to have markedly elevated urinary adrenaline concentrations, surprisingly catecholamine metabolites, both HMMA and total fractionated metanephrines were within their respective reference range. The laboratory was able to reproduce the exact biochemical picture by adding adrenaline (in the expected concentration based on the adrenaline content of her Anapen^®) to a normal urine collection. On occasions, the patient's urinary adrenaline concentrations were 10–20 times the upper limit of our reference range and clinically she reported symptoms of headache and palpitations, yet during her supervised hospital admission, her blood pressure and heart rate remained normal throughout.

Renal ultrasound imaging of the kidneys and non-contrast CT of the adrenals and pelvis were normal. I¹²³ MIBG scintigraphy was also normal. Normal radiological investigations with strikingly elevated urinary adrenaline concentrations in a clinically asymptomatic patient made adrenaline adulteration of urine collections the most likely explanation.

Her thyroid abnormalities suggested that she either had a transient thyroiditis, or she had taken exogenous thyroxine. The former is very unlikely given the low concentrations of inflammatory markers, lack of tenderness on examination of her neck and decreased uptake of thyroid isotope on scan.

This case was diagnostically very challenging, particularly as the patient appeared to be very credible. A close liaison between the clinicians, clinical scientists, general practitioner and pharmacist was crucial for the correct management of this patient.

With a greater access to medical knowledge through the Internet, patients are more likely to create diagnostic confusion for medical practitioners by having information about signs, symptoms and investigations of specific conditions. The most valuable learning point from this case is the need to remain vigilant and that a multidisciplinary team approach is vital when faced with difficult diagnostic dilemmas.