| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
CLINICAL CASE SEMINAR |
Departments of Endocrinology (L.J.H., M.E., S.L.C., J.P.M., K.A.M.) and Respiratory Medicine (J.A.W.), St. Bartholomew’s Hospital, London, United Kingdom EC1A 7BE; and Department of Medicine, Southend General Hospital (L.J.H., A.G.D., K.A.M.), Westcliff-on-Sea, United Kingdom SSO 0RY
Address all correspondence and requests for reprints to: Dr. K. A. Metcalfe, Department of Endocrinology, Southend General Hospital, Prittlewell Chase, Westcliff-on-Sea, United Kingdom SSO 0RY. E-mail: karl.metcalfe{at}southend.nhs.uk.
 |
Abstract |
|---|
 Top Abstract IntroductionCase Reports (summarized in...Materials and MethodsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionCase Reports (summarized in...Materials and MethodsDiscussionReferences |
|---|
OSA has a well documented association with increased cardiovascular morbidity and mortality (2, 3, 4, 5, 6, 7) and has been shown to be an independent risk factor for the development of systemic hypertension (8). Evidence is accumulating to suggest a role for sympathetic overactivity in the pathophysiology of these observations (9, 10, 11, 12, 13).
Pheochromocytomas may arise from anywhere in the autonomic nervous system, with 90% found in the adrenal medulla. Catecholamine-secreting tumors can also arise from extraadrenal neural crest derivatives. They are rare, but can be life-threatening. Presentation is typically with labile hypertension and symptoms of paroxysmal catecholamine excess (14). Once considered, the diagnosis is usually straightforward, because measurement of urinary catecholamines (15) provides a diagnostic screen. There are accumulating data to suggest the measurement of urinary metanephrines (16, 17, 18, 19) as a more sensitive biochemical screening test, but it remains the policy of our units, and at this point in time the majority of endocrinology departments, to use urinary catecholamines as the initial screening test for pheochromocytoma. Tumors are normally easily located by computerized tomography, magnetic resonance imaging, or radionuclide scanning with metaiodobenzylguanidine (MIBG) or octreotide (20). In the unusual event of negative imaging, investigation may proceed to selective venous sampling (21) or suppression tests with the ganglion blockers pentolinium (22) and clonidine (23).
We describe here a series of patients who have presented to us with clinical and biochemical features diagnostic of pheochromocytoma (Table 1
). Extensive investigations have not identified a catecholamine-secreting tumor in any of these patients. Subsequently, all have been diagnosed with OSA, which appears capable of causing massive sympathetic discharge and presentation as pseudo-pheochromocytoma in some individuals. This responds well to treatment with CPAP therapy. We suggest that a diagnosis of OSA should be considered in patients with clinical and biochemical evidence of catecholamine excess in whom a catecholamine-producing tumor cannot be identified.
|
|
Case Reports (summarized in Table 1 )
|
|---|
|
TopAbstractIntroductionCase Reports (summarized in...Materials and MethodsDiscussionReferences |
|---|
A 62-yr-old woman with morbid obesity and a past history of depression presented with panic attacks and hypertension of 204/100 mm Hg. Her doctor started her on nifedipine. She also complained of bilateral leg swelling and had a computed tomography (CT) scan of abdomen and pelvis to look for a compressive lesion, which revealed a 4 x 4-cm left adrenal mass resembling a cortical adenoma. Her urinary noradrenalines were elevated at 1030, 810, and 850 nmol/24 h, with normal adrenaline levels. Her cortisol dynamics were abnormal (elevated midnight cortisols of 419 and 296 nmol/liter; normal, <50 nmol/liter), but this was believed to be consistent with depression. MIBG scanning showed no uptake in the left adrenal mas, and 4 months later it was found to be unchanged in size on CT scan. However, the patient was hypoxic during an in-patient stay and had a sleep study consistent with OSA. After treatment with CPAP, her urinary noradrenalines normalized to 370 and 330 nmol/24 h, and systemic blood pressure fell to 135/75 mm Hg on venlafaxine, but no antihypertensive medication.
Case 2 1
A 53-yr-old woman presented with severe accelerated hypertension (200/110 mm Hg), grade 3 hypertensive retinopathy, and dyspnea. Prescribed drugs included perindopril, irbesartan, frusemide, lorazepam, and clomipramine. Urinary noradrenalines were elevated at 2145 and 1220 nmol/24 h. CT, MIBG, and octreotide scans failed to locate the suspected pheochromocytoma, and on venous sampling there was no reversal of the adrenal venous adrenaline to noradrenaline ratio and no noradrenaline hotspot. Plasma noradrenaline was elevated at 19.3 nmol/liter and did not suppress with pentolinium. Urinary and plasma adrenalines were normal throughout the study. A CT chest scan was performed because of dyspnea and hypoxia. This showed massive pulmonary arteries, with an estimated mean pulmonary arterial pressure at echocardiography of 80 mmHg. A sleep study was consistent with OSA. CPAP therapy was commenced, with rapid clinical improvement and normalization of urinary noradrenaline (556 nmol/24 h). Pulmonary arterial pressure fell to 39 mm Hg, and systemic blood pressure to 120/80 mm Hg. A repeat pentolinium suppression test showed basal and postpentolinium plasma noradrenaline within the normal range (3.3 and 3.0 nmol/liter). She has experienced adverse reactions to a variety of antihypertensive medications and is currently taking phenoxybenzamine, propranolol, frusemide, and spironolactone in addition to lorazepam and clomipramine as described above.
Case 3
A 42-yr-old man with type 1 neurofibromatosis had previously presented with a severe episode of hypertensive heart failure. At that time, urinary catecholamines were elevated (noradrenaline, 964 nmol/24 h; adrenaline, 236 nmol/24 h); CT and MIBG scanning showed a right adrenal phaeochromocytoma. After careful preparation with 
and ß blockade, this was successfully removed in an uncomplicated procedure, with initial normalization of systemic blood pressure and urinary catecholamines postoperatively. However, he soon presented again with resistant hypertension (160/110 mm Hg) and a persistently elevated urinary noradrenaline level (1324 nmol/24 h), but with normal urinary and plasma adrenaline levels. Medication at this time was amlodipine, frusemide, and perindopril. In a pentolinium suppression test, basal plasma noradrenaline was 6.4 nmol/liter; it was 6.5 nmol/liter 10 min after pentolinium administration. Repeat imaging by CT, MIBG, and octreotide scanning and venous sampling were all performed, and there was no further evidence of pheochromocytoma. On direct questioning he admitted to lifelong troublesome snoring. He then underwent a sleep study, which showed OSA. CPAP therapy was commenced. Systemic blood pressure fell to 135/84 mm Hg, and serial urinary noradrenaline levels normalized to 455 and 187 nmol/24 h on atenolol and perindopril. A repeat pentolinium suppression test during CPAP was also normalized (basal plasma noradrenaline, 2.3 nmol/liter; after pentolinium, 2.1 nmol/liter).
Case 4
A 48-yr-old man was referred by his doctor with hypertension of 170/100 mm Hg and severe headache unresponsive to treatment with bendrofluazide and atenolol. Urinary noradrenalines were raised at 870 and 814 nmol/24 h, with adrenaline within the normal range. Doxazosin was added in to his drug regimen with little effect on blood pressure. Imaging with CT and MIBG did not locate a pheochromocytoma. Venous sampling found a reversed adrenal vein noradrenaline to adrenaline ratio of 10:1 bilaterally, suggestive of occult pheochromocytoma. Sleep studies were consistent with moderate to severe OSA, and after CPAP therapy his blood pressure was 140/85 mm Hg, with a normal urinary noradrenaline level. His medications are otherwise unchanged.
Case 5
A 38-yr-old woman presented with severe headache and a blood pressure of 160/110 mm Hg. She was commenced on bendrofluazide by her doctor. Measurement of urinary catecholamines was variable over a period of 18 months, including several isolated elevated noradrenaline measurements of 830, 700, and 820 nmol/24 h. Blood pressure remained elevated despite the addition of atenolol and doxazosin. Her plasma catecholamine levels were normal. CT and MIBG scanning and venous sampling were also normal. Her sleep study showed moderate OSA, and after treatment with CPAP her urinary noradrenaline normalized to 450 nmol/24 h, with a blood pressure of 130/75 mm Hg and no alterations in drug therapy.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionCase Reports (summarized in...Materials and MethodsDiscussionReferences |
|---|
Blood samples were taken under standardized conditions, 30 min after venous cannulation, with the patient in a fixed resting position. Basal free catecholamines were estimated using HPLC (15). In suppression tests, after taking blood for basal catecholamine measurement, 2.5 mg pentolinium were given iv, and a repeat blood sample was drawn at 10 min, with monitoring of systemic blood pressure (18).
Urinary catecholamines
All urine was collected for 24 h, and catecholamines were determined by HPLC (15). For both plasma and urinary catecholamines, chromatograms were analyzed by laboratory staff according to protocols agreed upon by the pathology directorates in our two centers.
Sleep studies
These were carried out using a variety of sleep study apparatus to measure pulse rate, pulse oximetry, nasal air flow, and chest wall movement. OSA is defined as an apnea/hypopnea index (AHI) of more than five per hour; an apnea is a reduction in air flow by more than 50%, and a hyperpnea is a reduction in air flow by less than 50%, both for at least 10 sec. The AHI is the sum of the number of apneas plus the number of hyperpneas per hour. An AHI of 5–14/h is classed as mild OSA, 15–30 is moderate, and more than 30 is severe. CPAP is given at an AHI of 20 or more (24).
Consent
Appropriate informed consent was obtained from all patients at each stage of investigation and treatment.
|
Discussion |
|---|
|
TopAbstractIntroductionCase Reports (summarized in...Materials and MethodsDiscussionReferences |
|---|
It is noteworthy that in all of our patients, noradrenaline alone was consistently elevated. Only about 2% of plasma noradrenaline comes from the adrenal medulla under basal conditions; the vast majority is derived from the innervation of small arteries and arterioles. Measurement of excess urinary noradrenaline in these patients is likely to reflect neuronal, rather than adrenal, release and represent sympathetic nerve activity and synaptic overflow.
Since the 1980s it has been thought that sympathetic activity is up-regulated in patients with sleep-disordered breathing. Twenty-four-hour urinary catecholamine levels were initially found to be increased in patients with untreated sleep apnea compared with those with narcolepsy (12, 13). A small cohort of male patients with severe sleep apnea and elevated urinary catecholamine levels was found to lose the normal diurnal variation in sympathetic excretion, suggesting increased nocturnal sympathetic activity (25). These patients were all treated with tracheostomy, after which their catecholamine levels normalized. These studies and others showing similar findings have been criticized on the basis of their small size and the presence of several confounding factors, most notably hypertension. However, a recent population-based study of 2668 hypertensive males demonstrated increased levels of urinary metanephrine and normetanephrine in patients with OSA using analysis that controls for age, weight, blood pressure, and drug therapy (26). At the electrophysiological level, several studies have shown surges in sympathetic muscle nerve activity both during and immediately after episodes of acute apnea, a response that also occurs with arousal from sleep. Furthermore, resting sympathetic nerve activity during wakefulness is twice as high in apneic patients compared with sex-matched controls (9, 10, 11). CPAP therapy for OSA has been shown to reduce parameters of cardiac sympathetic tone (27, 28).
Exactly how this arises is widely debated. Animal studies have highlighted a role for both carotid body chemoreceptors and resetting of baroreceptor sensitivity (29). The fact that patients with hypertension and sleep apnea show a greater pressor response to hypoxia than patients with hypertension alone lends weight to the chemoreceptor hypothesis (30).
We hypothesize from this case series that in some patients with OSA the tendency for excessive catecholamine production can have significant cardiovascular consequences. Clinical improvement and normalization of the parameters of measurement of absolute catecholamine levels were seen in all cases after primary treatment of the OSA with CPAP.
Case 4 demonstrated a reversal of the adrenal vein noradrenaline:adrenaline ratio, a finding usually considered diagnostic of pheochromocytoma (21). This result was obtained at a time when systemic catecholamine levels were high. Subsequent diagnosis and treatment of OSA have resulted in clinical improvement and normalization of urinary catecholamine levels. It is not believed to be clinically appropriate to repeat this invasive procedure in the absence of further treatment implications. It does, of course, remain feasible that this patient and the others in our series harbor an occult pheochromocytoma despite rigorous negative imaging by CT, MIBG, and octreotide, and all patients continue under close monitoring.
We are obviously aware of recent studies suggesting that measurement of urinary fractionated metanephrines or plasma metanephrine has a higher sensitivity and specificity for the diagnosis of pheochromocytoma (16, 17, 18, 19). However, conflicting data come from a large group of hypertensive males, in whom urinary metanephrine, but not noradrenaline, was significantly elevated in those also suffering from OSA (26). At the time of clinical presentation of our patients, urinary noradrenaline and adrenaline was the initial screening test to investigate the possibility of pheochromocytoma, and clearly noradrenaline was significantly elevated in all of the patients described. Measurement of urinary metanephrines undoubtedly has an important role in the investigation of pheochromocytoma, but for the purposes of the cases described here it would have been unlikely to improve the sensitivity or specificity of the investigation strategies employed.
There is a well reported association between OSA and increased cardiovascular morbidity and mortality. This includes higher rates of angina and myocardial infarction, left ventricular hypertrophy and failure, hypertension, arrhythmias, and stroke. After years of debate, two robust epidemiological studies indicate that OSA is an independent risk factor for the development of hypertension (8, 31). As noradrenaline has a major effect on peripheral resistance, which is, in turn, a key determinant of systolic blood pressure, it is logical to suggest increased catecholamine levels as a factor in the development of hypertension in patients with OSA.
In the case series presented we hypothesize that OSA has caused massive sympathetic discharge and consequent presentation as pseudopheochromocytoma. The term pseudopheochromocytoma has been used to describe patients who have classical symptoms of pheochromocytoma in the presence of normal catecholamine levels (32). It is thought to be due to adrenergic hypersensitivity and is clearly quite different from the cases seen here. There are several recognized pharmaceutical causes of induced catecholamine excess, for example, cocaine (33), anti-Parkinsonian (34) and antipsychotic (35) medications, tricyclic antidepressants, and phenoxybenzamine (33). These produce a further sort of pseudopheochromocytoma (Table 2). Tricyclics can, in addition, profoundly suppress sympathetic nerve traffic and synaptic cleft overflow of noradrenaline (36), potentially further complicating the interpretation of catecholamine measurement and suppression tests in patients taking these drugs. One of our patients (case 2) was taking a tricyclic antidepressant at presentation and for many years previously, but has remained on this medication throughout treatment of her OSA and normalization of catecholamines, so her initial very elevated, nonsuppressible noradrenaline levels and accelerated hypertension would not appear to be drug-induced. Interestingly, she is now taking phenoxybenzamine, because it is one of the few antihypertensive drugs tolerated.
|
As OSA is widely prevalent in the adult population, it follows that some patients with true pheochromocytoma will have a positive sleep study. Clearly, all patients with OSA and elevated catecholamine levels should still be extensively investigated to exclude a catecholamine-secreting tumor. This case series would suggest, though, that in patients with catecholamine excess in whom such a tumor cannot be identified, OSA may be considered as a potential cause.
|
Footnotes |
|---|
1 Data from cases 2 and 3 have been published elsewhere as follows: Obstructive Sleep Apnoea and Pseudophaechromocytoma, Case Presentation at the Society for Endocrinology Clinical Practice Day, University of Reading, Reading, UK, July 12, 2002; and Obstructive Sleep Apnoea: A Cause of Pseudophaeochromocytoma, poster at the 193rd Society for Endocrinology Annual Meeting and Joint Meeting with Diabetes UK, London, UK, November 4–6 2002. The above is also published by the Society for Endocrinology as an abstract in Endocr Abstr 4:14–15, 2002. 
Received August 7, 2003.
Accepted February 18, 2004.
|
References |
|---|
|
TopAbstractIntroductionCase Reports (summarized in...Materials and MethodsDiscussionReferences |
|---|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
