Elevated urinary catecholamines and adrenal haemorrhage mimicking phaeochromocytoma

Abstract

A 51-year-old woman was admitted with left-sided flank pain initially thought to be renal colic. However, a CT urogram was normal. During the course of the admission the pain persisted and she developed severe sustained hypertension. A repeat CT scan of the abdomen revealed a 5×3 cm left adrenal abnormality consistent with haemorrhage, not seen on the original scan. Further assessment revealed elevated urine catecholamines and a short synacthen test showed a suboptimal cortisol response. The diagnosis was initially considered as a phaeochromocytoma, she received phenoxybenzamine with good resolution of hypertension and was referred for surgical opinion. However, serial urinary catecholamine concentrations returned to within the normal range and the diagnosis was revised to adrenal infarction and haemorrhage due to antiphospholipid syndrome. This case illustrates the importance of recognising adrenal infarction as a potential cause of ‘pseudophaeochromocytoma’.

Background

To highlight adrenal haemorrhage as a rare cause of hypertension and transient elevation of urinary catecholamines mimicking phaeochromocytoma.

Case presentation

A 51-year-old woman was admitted with a 1-day history of left-sided flank pain. She had no significant medical history, was not taking regular medications and had a family history of essential hypertension. Examination was unremarkable other than left-sided abdominal tenderness and blood pressure of 167/113 mm Hg. Urinalysis was positive for protein and blood. The only abnormalities noted on initial blood investigations included a modestly raised C reactive protein of 18 mg/litre (<6) and elevated white cell count of 15.5×10⁹ cells/litre. A diagnosis of urine tract infection or left renal colic was proposed, yet the CT urogram was reported as normal. She was treated with antibiotics but over the following week her flank pain deteriorated, being associated with episodic sweats, vomiting, a deterioration of hypertension (recorded at 205/125 mm Hg) together with a tachycardia of 120 beats/min. A repeat CT scan of the chest, abdomen and pelvis revealed lung airspace shadowing consistent with pulmonary oedema and a 5×3 cm hypodense abnormality of the left adrenal gland consistent with haemorrhage. This was not seen on review of the original CT urogram. She was transferred to the high dependency unit for monitoring and respiratory support. In view of the history and the presence of the adrenal mass, a diagnosis of infarction/haemorrhage within a phaeochromocytoma was considered and she was initially treated with phenoxybenzamine.

Subsequent investigation revealed an isolated elevation of normetepinephrine of 7.58 mol/24 h (<4.0) on 24 h urinary catecholamine assessment together with a suboptimal cortisol response (433 nmol/litre) 30 min after 250 g intravenous synacthen (normal response (NR) >550 nmol/litre). She was treated with intravenous hydrocortisone 50 mg three times a day together with increasing doses of phenoxybenzamine. Despite effective control of her blood pressure, her renal function declined with serum creatinine rising to a peak of 517 mol/litre (NR 50–100). Further investigations revealed a negative antiglomerular basement membrane antibody. A clotting profile showed an elevated activated partial thromboplastin time of 54.7 s (22.0–31.5), anti-cardiolipin antibodies were positive and other haematological indices were consistent lupus anticoagulant (table 1). Antiphospholipid syndrome was therefore suggested as a cause for the haemorrhage of a possible underlying phaeochromocytoma. She underwent renal biopsy which reported widespread ischaemic change and small vessel hyalinosis. No definite microthrombi were seen but appearances were suggestive of resolving thrombosis.

Table 1.

Results of haematological and immunological indices

	Result	Normal Range
APTT (Patient)	54.7 s	(22.0–31.5)
Russell Viper Venom Ratio		3.16	(0.78–1.22)
Lupus sensitive APTT	66.6 s	(23.0–33.0)
Dilute Russell Viper Venom Test pat corr 20% Normal plasma	137.4 s
Dilute Russell Viper Venom Test corr 20% Normal plasma	3.12
Consistent with lupus anticoagulant
IgG b2 glycoprotein	8.9 U/ml	(0.0–5.0)
IgM b2 glycoprotein	9.6 U/ml	(0.0–5.0)
Anti-b2GP1 antibodies are thought to be more closely associated than anti-cardiolipin antibodies with thrombosis in patients with a history of anti-phospholipid antibody-associated disease
IgG cardiolipin Abs	15 G-phospholipid units/ml	(<10)
IgM cardiolipin Abs	22 M-phospholipid units/ml	(<7)
Anti-nuclear Abs	Negative
Crithidia dsDNA	Negative
Anti-dsDNA Abs	37 IU/ml	(<20)
The negative ANA and crithidia anti-DNA immunofluorescence tests indicate the anti-dsDNA ELISA result is a false positive due to non-specific binding.
Extractable Nuclear Antigen screen	Negative

Abs, antibodies; ANA, anti-nuclear antibodies; APTT, activated partial thromboplastin time; dsDNA, double-stranded DNA.

Anti-b2GP1 antibodies are thought to be more closely associated than anti-cardiolipin antibodies with thrombosis in patients with a history of anti-phospholipid antibody-associated disease.

The negative ANA and crithidia anti-DNA immunofluorescence tests indicate the anti-dsDNA ELISA result is a false positive due to non-specific binding.

She was referred for an endocrine opinion for management of what was considered to be an underlying phaeochromocytoma. However, the diagnosis was reconsidered in the light of no previous suggestive symptoms and the initially normal CT urogram. The diagnosis was felt to be pseudophaeochromocytoma due to adrenal infarction and haemorrhage associated with the antiphospholipid syndrome. This was supported by a decline and normalisation of urine catecholamine concentrations during hospitalisation (table 2).

Table 2.

Serial 24 h urine catecholamine concentrations

	Initial	3 days	12 days	28 days	34 days
Urine normetepinephrine/24 h umol/24 h (<4)	7.58	5.1	8.57	1.87	0.72
Urine metetepinephrine/24 h umol/24 h (<2)	0.74	0.2	0.82	0.07	0.04

Over the course of the admission phenoxybenzamine was changed to amlodipine and subsequently her blood pressure improved to 104/70 mm Hg at which time antihypertensives were withdrawn. Renal function continued to improve with a serum creatinine 244 µmol/litre at discharge. A repeat short synacthen test showed a persistent suboptimal adrenal response with a cortisol rising to 215 nmol/litre at 30 min after intravenous synacthen. She was maintained on hydrocortisone 20 mg morning and 10 mg evening daily and given fludrocortisone 100 g once daily. At discharge, 6 weeks after admission, urinary catecholamines had normalised (table 2).

Differential diagnosis

• ▶Adrenal apoplexy
• ▶Malignant hypertension
• ▶Pseudophaeochromocytoma
• ▶Phaeochromocytoma
• ▶Antiphospholipid syndrome

Treatment

• ▶Hydrocortisone 20 mg morning, 10 mg evening
• ▶Fludrocortisone 100 μg per day

Outcome and follow-up

The patient is well and remains under outpatient follow-up. She continues with glucocorticoid and mineralocorticoid treatment and awaits reassessment of adrenal reserve. She remains normotensive without any antihypertensive treatment. Renal function has improved with a creatinine level of 165 mol/litre (50–100). As yet, she has not received formal anticoagulation. A repeat ultrasound scan of the left adrenal gland shows the mass smaller at 3.9×2.3×3.4 cm.

Discussion

Phaeochromocytoma presents with a constellation of diverse symptoms ranging from classical palpitations headaches and sweats to abdominal pain.¹ Our case illustrates the rarely described event of adrenal infarction/haemorrhage mimicking a phaeochromocytoma, so-called pseudophaeochromocytoma.

Measurement of urinary catecholamine and metabolite concentrations is a very useful initial screening tool for phaeochromocytoma with reports indicating sensitivity and specificity of 87.5% and 99.7%, respectively.² However, false positive results are reported, being described in conjunction with the use of drugs such as labetolol together with foods such as fruits and nuts.^{3 4} Rarely, elevated catecholamine concentrations are reported in association with adrenal haemorrhage/infarction and in some cases this has prompted surgical removal of the gland for suspected phaeochromocytoma.^{5 6} In our case, phaeochromocytoma was postulated as a diagnosis and surgical opinion was sought. However, reference to the initial CT scan appearance, the absence of any prior symptoms and the appearance of the apoplectic adrenal gland with resolution of serial catecholamine concentrations refuted the diagnosis and averted unnecessary surgery.

Another possible explanation for falsely elevated urine catecholamine concentrations is acute illness. Our patient presented with acute pain and rapid clinical deterioration resulting in intensive care unit (ITU) admission. Indeed, we acknowledge that her acute illness is likely to have contributed to an increase in catecholamine concentrations; mild elevations of urinary catecholamines are reported during acute stress⁷ and in the ITU setting.⁸ This may well have contributed to her presentation with ‘pseudophaeochromocytoma’. It may be argued that our case represents apoplexy of a pre-existent phaeochromocytoma and that subsequent normalisation of catecholamine concentrations was a consequence of autoinfarction of the tumour. However, the correct diagnosis is supported by the absence of any prior symptoms and the initially normal CT urogram. In this case, it was important to make the correct diagnosis, as unnecessary surgery was avoided.

Adrenal infarction is rare and underlying diagnoses should be considered. In our patient, the presence of prolonged activated partial thromboplastin time (APTT), presence of anti-cardiolipin antibodies together with the presence of adrenal infarction supports a diagnosis of the antiphospholipid syndrome. Adrenal infarction is a recognised sequelae⁹ and it is postulated that this occurs as a consequence of adrenal vein thrombosis. The adrenal gland is highly vascular, supplied with up to 50 small arterial branches originating from the inferior phrenic, renal arteries and the aorta. However, blood drains through relatively few venous channels before drainage via a single adrenal vein. Thus, the adrenal vein is prone to venous thrombosis due venoconstriction following the surge of catecholamines released at times of stress particularly in a procoagulant environment.¹⁰

It is interesting to note that although the adrenal apoplexy was unilateral, our patient developed adrenal insufficiency that persisted following recovery from the acute illness. Although adrenal insufficiency is recognised in acute illness,¹¹ its persistence suggests that the prothrombotic environment produced bilateral adrenal infarction accompanied by unilateral adrenal haemorrhage. In support of this conclusion is the suggestion of microthromboses on renal biopsy.

In summary, this case illustrates the rarely recognised event of adrenal apoplexy presenting as a pseudophaeochromocytoma with hypertension, tachycardia and elevated urine catecholamine concentrations. Careful evaluation with review of the symptoms, investigations and serial decline of urine catecholamine concentrations permitted exclusion of a phaeochromocytoma and avoided surgical exploration. While the measurement of urine catecholamine concentrations is an effective screening tool for the diagnosis of phaeochromocytoma, it is important to appreciate the causes of falsely positive results.

Learning points.

• ▶Phaeochromocytoma may present with many diverse symptoms and signs.
• ▶The 24 h urine catecholamine findings are a good screening test for phaeochromocytoma.
• ▶Doctors should appreciate the causes of ‘psuedophaeochromocytoma’.
• ▶Adrenal apoplexy is a presenting feature of antiphospholipid syndrome.