www.cherylswoffordnets.com
Email: cherylswoffordnets@yahoo.com
Email: cherylswoffordnets@yahoo.com
NEUROENDOCRINE TUMOR HYPOTHESIS
OF MENTAL AND PHYSICAL DISEASES
This website is devoted to presenting my neuroendocrine
hypothesis of disease. I am an independent researcher in
this field, working in Atlanta, Georgia, USA as of January
2007. My hypothesis suggests that neuroendocrine tumors
(NETS), such as pheochromocytomas, paragangliomas, and
other such tumors that secrete neuro-transmitters such as
epinephrine (or adrenaline), norepinephrine (or
noradrenaline), and dopamine, are a vastly under-diagnosed
possible cause of many types of problems, such as
schizophrenia, anxiety, hypertension, sleep apnea, sudden
infant death, and many other behavioral, mental and physical
conditions or disease states.
Cheryl D. Swofford MD,PsyD
OF MENTAL AND PHYSICAL DISEASES
This website is devoted to presenting my neuroendocrine
hypothesis of disease. I am an independent researcher in
this field, working in Atlanta, Georgia, USA as of January
2007. My hypothesis suggests that neuroendocrine tumors
(NETS), such as pheochromocytomas, paragangliomas, and
other such tumors that secrete neuro-transmitters such as
epinephrine (or adrenaline), norepinephrine (or
noradrenaline), and dopamine, are a vastly under-diagnosed
possible cause of many types of problems, such as
schizophrenia, anxiety, hypertension, sleep apnea, sudden
infant death, and many other behavioral, mental and physical
conditions or disease states.
Cheryl D. Swofford MD,PsyD
 | Research report below: |
| My schizophrenia hypothesis is a subset of my |
I plan to eventually further investigate the possible
relationship with NETS and other diseases
| You may read the 1-page abstract directly below for a summary |
The pheochromocytoma hypothesis of schizophrenia
Abstract
This paper’s focus is the hypothesis that pheochromocytomas are a non-diagnosed major cause of schizophrenia
because they secrete neurotransmitters, such as dopamine. The author’s neuroendocrine tumor (NET) hypothesis
of disease is briefly presented, suggesting NETs as an under-diagnosed and non-diagnosed cause of many cases
of behavioral, mental, and physical conditions.
Statistics of 1% of 30,000 persons having adrenal tumors (the number of pheochromocytomas is not known)
in a review of fifty years of reports of autopsies or imaging (Allolio, 2001) suggest the possibility of a much
greater prevalence of pheochromocytomas than some reports of a 0.3% prevalence (Sheps, Jiang et al, 1990),
found mostly in hypertensives. The widely stated statistic that an estimated 1% of persons worldwide has
schizophrenia is presented, perhaps of significance to the pheochromocytoma hypothesis. The fact that at least
some pheochromocytomas secrete dopamine, that they can cause psychosis, and that the dopamine hypothesis
prevails for schizophrenia are used to support the paper’s primary hypothesis.
The types, possible causes, and clinical manifestations of pheochromocytomas are named. Findings of research
suggesting vast under-diagnosing (of >99%) are presented. Symptom variability, such as having hypertension or
not, is explained as a factor in under-diagnosing. Testing modalities and their sensitivities and specificities are
covered. Cost of testing and implications for not testing are presented. Questions are raised as to possible
significant effects of under-diagnosed pheochromocytomas and other NETs.
Abstract
This paper’s focus is the hypothesis that pheochromocytomas are a non-diagnosed major cause of schizophrenia
because they secrete neurotransmitters, such as dopamine. The author’s neuroendocrine tumor (NET) hypothesis
of disease is briefly presented, suggesting NETs as an under-diagnosed and non-diagnosed cause of many cases
of behavioral, mental, and physical conditions.
Statistics of 1% of 30,000 persons having adrenal tumors (the number of pheochromocytomas is not known)
in a review of fifty years of reports of autopsies or imaging (Allolio, 2001) suggest the possibility of a much
greater prevalence of pheochromocytomas than some reports of a 0.3% prevalence (Sheps, Jiang et al, 1990),
found mostly in hypertensives. The widely stated statistic that an estimated 1% of persons worldwide has
schizophrenia is presented, perhaps of significance to the pheochromocytoma hypothesis. The fact that at least
some pheochromocytomas secrete dopamine, that they can cause psychosis, and that the dopamine hypothesis
prevails for schizophrenia are used to support the paper’s primary hypothesis.
The types, possible causes, and clinical manifestations of pheochromocytomas are named. Findings of research
suggesting vast under-diagnosing (of >99%) are presented. Symptom variability, such as having hypertension or
not, is explained as a factor in under-diagnosing. Testing modalities and their sensitivities and specificities are
covered. Cost of testing and implications for not testing are presented. Questions are raised as to possible
significant effects of under-diagnosed pheochromocytomas and other NETs.
The pheochromocytoma hypothesis of schizophrenia,
A subhypothesis of my NETS (neuroendocrine tumors) hypothesis of disease
[NETS and behavioral, mental, and physical disorders]
For Neuroendocrine tumor/Pheochromocytoma
website www.cherylswoffordnets.com
Author: Cheryl D. Swofford, MD,PsyD,MSci,MArts
Atlanta, Georgia and Marion, NC, USA
Last update: January 26, 2007
Word counts: Abstract: 175 Text: 4000
Keywords: neuroendocrine tumors, NETS, pheochromocytoma, schizophrenia, hypertension,
anxiety, paraganglioma, psychosis, adrenal tumor, chromaffin cell tumor
“The miniature structure itself protrudes up through the floor as though it were the tip of an iceberg—
the apex of an enormous, pyramidical vault, submerged below like a hidden chamber.”
Dan Brown. In The Da Vinci Code (p. 488).
I hypothesize that pheochromocytomas (and paragangliomas) are a non-diagnosed cause of many cases of
psychiatric disorders, in particular anxiety, schizophrenia and other psychotic disorders. Further, I hypothesize
that neuroendocrine tumors (NETs) (of which pheochromocytomas and paragangliomas are two of many
different types) are a non-diagnosed cause of a significant number of cases of mental and physical conditions
[such as, asthma, benign prostatic hypertrophy or hyperplasia (BPH), blindness, cancers, cardiac disorders,
cranial neuropathy, headache, seizures, and other neurological disorders, depression, diabetes, hearing loss,
hypoglycemia, hyperthyroidism and other endocrine disorders, eclampsia/pre-eclampsia of pregnancy,
homosexuality, hyperactivity, hypertension, indigestion, menopausal symptoms, Raynaud’s, sudden infant death
syndrome, sleep apnea and other sleep disorders, etc.] However, NETs other than pheochromocytomas and
paragangliomas, as well as non-psychiatric disorders, are beyond the scope of this paper. I plan to address
those topics in future papers and a book.
Pheochromocytomas and paragangliomas are the two subtypes of chromaffin cell tumors. Pheochromo-
cytomas come from the adrenal medulla, and paragangliomas are extra-adrenal, derived from sympathetic or
parasympathetic paraganglia. A widely-recognized “rule of 10” roughly applies to chromaffin cell tumors:
10% extra-adrenal, 10% malignant, 10% familial, 10% not associated with hypertension, 10% bilateral, and 10%
in children (Kaltsas, Besser & Grossman, 2004). The remainder of this paper will refer to pheochromocytomas
primarily, since they are thought to account for about 90% of these tumors. [However, keep in mind that these
tumors are vastly under-diagnosed and we really do not know the numbers.] [Paragangliomas have been found
in the heart, along the para-aortic sympathetic chain, at the organ of Zuckerkandl (at the aortic bifurcation), in the
neck (carotid body tumor or glomus tumor), the mediastinum, and in and around the urinary bladder. They can
exist anywhere there are chromaffin cells. Less than 1% of chromaffin cell tumors have been found to be extra-
abdominal (Benowitz, 1988).
Pheochromocytomas are known for their sympathetic nervous system (SNS) manifestations, due to their
release of neurotransmitters, the catecholamines epinephrine (or adrenaline) and norepinephrine (or noradren-
aline). The tumors also release the catecholamine dopamine in lesser amounts and fewer cases, it is thought.
I hypothesize that any bodily function affected by the SNS and catecholamines has the potential of being affected
by pheochromocytomas. The primary focus of this article will be schizophrenia and its possible connection with
dopamine, with the underlying assumption that psychotic disorders in general may be related to pheochromo-
cytomas. Anxiety and other disorders will also be mentioned.
Types
Pheochromocytomas may be genetic or sporadic, local and “benign,” non-malignant or malignant, non-
metastatic or metastatic. [To say that they are sporadic to suggest they are not genetic may not be accurate.
To say they are “benign” because they are not malignant is more or less a misnomer.] They may go undiagnosed
and cause few to many problems. They may be diagnosed and an association may not be made with all the
problems they have caused. They may be diagnosed and cause variable problems in an unpredictable, varying
manner.
Causes
The causes of most pheochromocytomas, other than perhaps metastatic tumors, are unknown. Most are
thought to be caused by genetic mutations.3 An estimated 10-20% of identified pheochromocytomas are found
in patients with four well-known disorders due to gene mutations causing an autosomal dominant pattern of
inheritance: multiple endocrine neoplasia syndromes (MEN 2A and 2B), von Hippel-Lindau disease (VHL) and
neurofibromatosis type 1 (NF1).
Slawik and Reincke (2003) report that “Adrenal nodularity increases with age and usually affects both glands”
(p. 17), leading to the hypothesis of aging as a cause of adrenal tumors. They explain, however, that most of the
tumors have a monoclonal origin, the result of an oncogenic mutation and that a lesser percentage of cases result
from polyclonal nodular hyperplasia in aging. They estimate that 75% of nonadrenal malignant tumors metastasize
to the adrenal glands (25-100%, depending on the type of cancer). This suggests a highly significant interaction
with the adrenals and other bodily systems and, possibly, with pheochromocytomas as well.
Although pheochromocytomas may not be a disease of aging, necessarily, age can be a factor. We are more
prone to heart disease, cancers and other tumors, etc as we age. For example, some paragangliomas, such as
carotid body tumors, with possible links to sleep apnea, and respiratory and cardiac problems, have been found to
have a peak incidence at ages 45-50 (Emedicine, 2005). [Since many persons in our society also start having
problems around that age thought to be due to aging and genetics and lifestyle, such as hypertension and heart
disease, it may be unlikely that a tumor is suspected, when it could be causing serious problems.]
Clinical manifestations
Clinical manifestations of diagnosed pheochromocytomas are primarily severe headache, palpitations,
profuse sweating, and hypertension. All diagnosed (with pheochromocytoma) patients have not had all of these
signs or symptoms.
Pheochromocytomas can also cause psychosis, paranoia, anxiety, and confusion; they do not in every case,
or even in most cases, as far as is known. But I hypothesize that pheochromocytomas may be a major cause
and at least a more significant cause of the aforementioned disorders than is now recognized. Much is still not
known about the cases they cause.
In our age of political correctness, and with the stigma attached to certain disorders over many years, it is
highly possible that studies that examined pheochromocytomas failed to report whether any of the patients had
a psychotic disorder or some other psychiatric disorder. [I have seen many charts of patients diagnosed with
schizophrenia or HIV, for example, with non-psychiatry or non-infectious disease physician notes that made little
or no mention of these diagnoses, even when it could have been relevant to the case.] Patients presenting with
psychiatric disorders by psychiatrists and non-psychiatrists generally are not tested for pheochromocytomas.
Non-psychiatric physicians generally are not looking for the cause of schizophrenia at all.
Pheochromocytomas have also been linked with causing vasomotor changes, facial pallor, bradycardia,
tachycardia, precordial pain, abdominal pain, anxiety, nervousness, irritability, weight loss, increased appetite,
angina, hypertension, cardiac enlargement, postural hypotension, elevated temperature, retinal and cerebro-
vascular hemorrhage, sudden blindness, seizures, hyperglycemia, paresthesias, Raynaud’s, pulmonary edema,
heart failure, cardiomyopathy, Cushing’s syndrome, erythrocytosis, hypercalcemia, hypermetabolism (with
normal thyroid function tests), thyrotoxicosis, hyperglycemia, etc (Tierney et al, eds., 2000). Invasive diagnostic
procedures and/or surgery in patients with pheochromocytomas that have not been diagnosed, can lead to severe
hypotension and shock, renal failure, and myocardial infarction (Tierney et al, eds, 2000).
According to Benowitz (1988) “Abnormal carbohydrate metabolism, particularly the sudden, unexpected
development of diabetes mellitus, should suggest a pheochromocytoma” (p. 564). Also, “Unusual hypertensive
responses to anesthesia, to any antihypertensive drugs (particularly beta-blockers), or to sympathomimetic
drugs are useful clues to the presence of the tumor” (p. 564). Patients in a hypercalcemic state may also exhibit
symptoms caused by calcium as a trigger for the release of catecholamines by a pheochromocytoma (Benowitz,
1988).
There is no reason to think that many of these signs or symptoms have a greater incidence in a schizo-
phrenia patient population than in the general patient population. However, it is known that at least one of the
above symptoms, psychosis, has a greater incidence among persons with schizophrenia. So why not examine
pheochromocytomas in more detail in schizophrenia patients? We need to keep in mind, however, that the use
of additional medications among schizophrenia patients vastly complicates getting a clear picture of possible
pheochromocytoma effects. There is likely a huge number of patients diagnosed with schizophrenia today,
started on antipsychotic drugs, and not adequately examined and tested for pheochromocytomas or not tested
prior to starting new drugs that could influence test results.
Under-diagnosing
Pheochromocytomas may be involved in any of the preceding problems but not diagnosed. A 1990
(Sheps, Jiang et al) report of data from Mayo Clinic indicates that only 1-2 of 300 out of 100,000 persons with
pheochromocytomas at autopsy are diagnosed in life (0.3% incidence of pheochromocytomas in autopsy cases,
with more than 99% non-diagnosed). [Mayo Clinic started keeping data on adrenal tumors when Dr. C. H. Mayo
removed such a tumor in 1926 (Sheps et al, 1990).] Allolio (2001) found a 1% prevalence of adrenal tumors in
more than 30,000 patients in autopsy series and computed tomography (CT) studies from 1944-1994. This
finding strongly suggests the need for imaging such as CT to detect the tumors.
Pheochromocytomas have been widely missed and “misdiagnosed as…anxiety, tension states, psycho-
neurosis, psychosis, hypoglycemia, menopause, headaches [including migraines and other types],…drug abuse:
amphetamines, cocaine, LSD, phenylpropanolamine, caffeine,…interaction of monoamine oxidase inhibitors with
certain food and beverages,…clonidine withdrawal,…atropine administration, [use of] nasal decongestants [and]
tricyclic antidepressants,…trauma, tumor, stroke, focal arterial insufficiency, autonomic hyperreflexia,…seizure,
…angina pectoris, exaggerated response to hypovolemia or infection,…tachycardia, familial dysautonomia
(Jacomet, 2001). Interestingly, a recent paper reported a case in which pheochromocytoma was found where
relapsing paranoid psychosis was described as the first manifestation of the tumor (Benabarre et al, 2005).
Are we onto something of much greater significance here, not yet revealed because of all the currently missed
cases?
Pheochromocytomas are considered rare, and statistics are largely limited to co-occurrence with hyper-
tension, occurring in an estimated one to five per 1000 cases of hypertension (NCIPDQ 2006). Given that an
estimated 20% of Americans have hypertension, and an estimated 1% of Americans [and persons worldwide]
(Harvard Consumer Health Information, 2006) has schizophrenia, we should be looking more closely at
pheochromocytomas as a cause, starting with all schizophrenia patients with uncontrolled hypertension and
expanding our search as much as possible. Ideally, we should not limit testing to hypertensives, however, given
the symptom variability, mentioned later.
Prior to starting this paper, I was in a position to partially examine each of more than 10,000 cardiology
patient charts, many for patients with hypertension and/or other cardiovascular problems. I read the most
recent assessment and plan in all of these patient charts. However, I recall fewer than five cases in which
testing for pheochromocytoma was mentioned. I worked in a department of psychiatry at a major university
teaching hospital for several years, working with schizophrenia researchers. I did not have pheochromo-
cytomas in mind and was not doing chart reviews related to pheochromocytomas. I do not recall anyone
mention pheochromocytomas as a possible cause of psychiatric disorders. The physicians and other
clinicians and researchers were experts in their fields, competent and respected. This is not to suggest
that they were not doing their jobs by current standards. I am suggesting that further testing for
pheochromocytomas, especially among hypertensive and schizophrenia patient populations, could reveal
more insight into the role of pheochromocytomas in hypertension as well as schizophrenia and other mental
and physical disorders.
Dopamine
An estimated 10% of pheochromocytomas are malignant. Malignant pheochromocytomas appear more
likely to produce and secrete more dopamine. According to Pacak (2002) elevated plasma or urinary dopamine,
“when accompanied by elevations in plasma norepinephrine or other clinical evidence of pheochromocytoma…
should arouse immediate suspicion of metastatic disease” (p. 6). An estimated 10-15% of persons with
schizophrenia die by suicide. The factors involved in the suicide of a person with schizophrenia are complex,
needless to say. Is it possible, though, that among the most ill patients with schizophrenia are those with
malignant, dopamine-secreting pheochromcytomas, and they die by suicide without being diagnosed? Since
we do not have the answers to these questions, we need to be seeking them. We now have the means.
It was not until the 1980s that dopamine release by pheochromocytomas became recognized in medical and
psychiatry circles. As recently as 1986, Proye et al (1986) published a paper questioning dopamine as an
unrecognized entity in pheochromocytomas. Yet, the possibility of pheochromocytomas as a significant cause
of psychotic illnesses has not been explored, although the dopamine hypothesis of psychotic illness continues.
The dopamine hypothesis of schizophrenia never goes away. Early work related to this hypothesis came
about when it was found that drugs that block brain dopamine function reduced psychotic symptoms.
[Near the mid-twentieth century chlorpromazine was used to control blood pressure elevated by certain
anesthesias used during surgeries. Then people began to notice that chlorpromazine reduced psychosis in
psychotic surgery patients; thus the advent of antipsychotic drugs. The connection with dopamine, hypertension,
and pheochromocytomas will not go away, either. Many anti-psychotic drugs can cause both hypotension and
neuroleptic malignant syndrome, with malignant hypertension. This subject is not simple.]
The brain has been studied in nearly every way imaginable. Still, the cause of schizophrenia and other
psychotic disorders is a mystery. However, evidence of excess brain dopamine remains. My hypothesis is that
we need to look outside the brain for the source of excess dopamine, to wit, pheochromocytomas and possibly
other NETs.
There is strong evidence of a correlation between peripheral dopamine and schizophrenia.
Ilani, Ben-Shachar et al (2001), with Professor Sara Fuchs, say that “Although the precise pathophysiology of
schizophrenia remains unknown, the dopaminergic hypothesis of the illness assumes that the illness results from
excessive activity at dopamine synapses in the brain” (p 625). They looked beyond the brain. They found
evidence of increased levels of peripheral dopamine (the D3 subtype) in schizophrenia patients, with
“a significant elevation of at least 2-fold in the mRNA level of the D3, but not the D4, dopamine receptor in
schizophrenic patients” (p. 625). They specifically looked at dopamine receptors on peripheral blood
lymphocytes, as did Kwak, Koo et al (2001) around the same time.
Pheochromocytomas have been found at autopsy in babies from two days old to adults 92 years old
(Sutton et al, 1981). Psychotic disorders have been diagnosed in children as well as adults (so has hypertension).
A study of home movies of children indicated that viewers accurately identified those children who later developed
schizophrenia (Walker & Lewine, 1990). Pheochromocytomas can enlarge with age, possibly growing larger
earlier in males, perhaps in a pattern consistent with schizophrenia. Much is going on that has not been explained
adequately. Let’s look at pheochromocytomas.
We know that pheochromocytomas can “masquerad[e] for decades” (Boorman & Erie Vann, 2000, p 2), as in
the case of a 38-year-old man who, in retrospect, had bizarre, hypervigilant, hyperactive behavior in high school.
He was found to have an elevated level of norepinephrine when diagnosed with a “benign” 3 cm pheochromocytoma
shown by an MRI. “His world had been veiled by the misperceptions caused by the pheochromocytoma, but the
veil was removed” (p 2). How many children may be hyperactive and how many adults may have anxiety because
they have a pheochromocytoma that no one knows about? We have no idea. There may be other tumors producing
dopamine and causing psychosis, or epinephrine or other chemicals causing anxiety and other disorders that no
one knows about. How many people are out there suffering, with tumors secretly secreting chemicals that are
negatively affecting them in all sorts of ways, behaviorally, mentally, and physically? We have the tools and we
could find out.
Symptom variability
According to Benowitz (1988) “the pattern of catecholamine release and related clinical manifestations [are]
variable” (p 561) in pheochromocytomas. The common symptoms, as we know them, of pheochromocytomas,
spells of severe headache, sweating, and palpitations, occur in about 90% of diagnosed patients. [One can see
that such patients might be misdiagnosed with anxiety, or have anxiety due to a pheochromocytoma.] But what are
the symptoms in the estimated greater than 99% of patients who are not diagnosed? Patients do not necessarily
exhibit all three of the most commonly known symptoms, but having these three symptoms together is considered a
strong diagnostic clue. There can be opposite symptoms and disagreement about symptoms. For example, pallor is
considered common, but some books will say flushing is a common symptom; others say flushing is rare
(Benowitz, 1988). Bradycardia is a common reflex response to norepinephrine-related vasoconstriction. But
tachycardia is perhaps a less common effect of epinephrine released by these tumors.
Hypertension, hypermetabolism, and weight loss are other commonly known manifestations. Dopamine and
epinephrine-producing tumors have, however, been found with a hypermetabolic state and normal blood pressures
(Benowitz, 1988). More than 90% of persons diagnosed with pheochromocytomas have hypertension (Benowitz,
1988). However, fifty years of Mayo Clinic data found hypertension in only 54% of patients diagnosed with
pheochromocytoma at autopsy (Sutton et al, 1981). I hypothesize that pheochromocytomas are mainly diagnosed
only in persons with hypertension and largely non-diagnosed in those who do not have hypertension or have
hypertension that is considered idiopathic, “essential” hypertension without testing for the tumors. Since
hypertension and the most commonly known manifestations of pheochromocytomas are not the only signs and
symptoms, we need to be looking far beyond those patients for pheochromocytomas.
Bravo et al (1979) did not find a strong correlation with circulating catecholamine levels and blood pressure in
pheochromocytoma, as one might expect. I hypothesize that the lack of correlation may also apply to other clinical
manifestations of pheochromocytomas. Such symptom variability would reduce suspicion of pheochromocytomas
and suggests a strong likelihood of under-diagnosing the tumors, perhaps a much greater likelihood than we might
suspect. Benowitz (1988), citing a variety of research, explains some of the reasons for such variability, related to
enzyme activity, hypovolemia, the release of both vasoconstrictor and vasodilator substances by the tumors, down-
regulation, sympathetic neural control of blood pressure, and such factors as change in posture, exercise, and
emotional state. The physiology is quite complex, causing confusion, misunderstanding, and vast under-diagnosing.
However, we have the diagnostic tools for rather accurate detection if we have the wherewithal.
Testing
Diagnostic testing modalities for pheochromocytomas have advanced greatly in recent years. Physicians/
scientists are still under-testing for pheochromocytomas and are still missing a large of percentage of cases.
Physicians are misdiagnosing a great number of cases of diseases heretofore not recognized as being related to
pheochromocytomas and that can be treated by reducing or removing tumors. Pheochromocytomas are still
considered rare, but they may be much more common than we know. Testing for them, and adequate testing, may
be what is rare. One physician recently told me, “Pheos are a pain to diagnose, and you have to keep testing and
testing.”
When testing is performed it often involves collecting 24-hour urinary vanillylmandelic acid (VMA),
catecholamines and metanephrines. [Catechol-o-methyltransferase (COMT) and monoamine oxidase (MAO) are
the two major enzymes that metabolize catecholamines to produce VMA; metanephrines are break-down products
of epinephrine.] Such urine testing can result in a “false negative” if the person being tested did not have a recent
exacerbation of symptoms related to catecholamine release or did not follow instructions. Instructions are to avoid
alcohol, coffee, tea, tobacco, and strenuous exercise for three days prior to testing. The same applies to blood
testing, plus at least eight hours of fasting (with only water), and either lying supine for 30 minutes or sitting upright
for 15 minutes are required.
Antipsychotic and antidepressant drugs and any drugs that influence the sympathetic nervous system can
affect results. Benowitz (1988) points out that calcium channel blockers can decrease “urinary catecholamine
secretion in patients with pheochromocytoma” (p. 566). Pacak (2002) recommends stopping acetaminophen for
at least 5 days prior to blood testing for pheochromocytomas. Pacak further points out that impaired renal function
(as in patients with congestive heart failure and hypertension) can cause high values for metabolic end-products
that can be misinterpreted as a pheochromocytoma.
Benowitz (1988, p 564) says that “Assays of urinary catecholamines and metanephrines are… adequate to
diagnose pheochromocytoma in most patients, but when the results of these studies are equivocal and
pheochromocytoma is suggested clinically, plasma catecholamines should be measured.” “There may be
considerable swings in the levels of plasma catecholamines from time to time, a consequence of postural changes,
exercise, emotional arousal, and other factors. Despite such variability, basal concentrations are usually increased
severalfold in pts with pheochromocytoma.” “To diagnose a dopamine-secreting pheochromocytoma, assays of
urinary dopamine may be much more sensitive.”
Imaging techniques used to locate pheochromocytomas are computed tomography (CT) scan, magnetic
resonance imaging (MRI), and metaiodobenzylguanidine (MIBG) scan. MRI has the advantages of avoiding IV
contrast material and irradiation and better differentiating tissue than CT. MIBG (or I-MIBG), using radioactive iodine
(usually 131I in the United States) is more sensitive and specific and can detect better than CT or MRI whether a
mass is a functioning pheochromocytoma or a benign adenoma.
Urine testing is generally considered more useful than blood testing for pheochromocytomas. That is not
necessarily true; it appears to depend on the type of test and type of pheochromocytoma, whether it is genetic or
sporadic, the age of the patients, and what the tumor is secreting when. Testing of 24-hour urine provides
“acceptable sensitivity and significantly better specificity than fractionated plasma free metanephrine,” per
Kudva et al (2003) in cases of sporadic pheochromocytoma presenting with triggers such as resistant hyper-
tension and spells (headache, profuse sweating, and tachycardia). However, Pacak (2002) recommends
fractionated plasma free metanephrine testing in the young pediatrics population for practical purposes.
Pacak (2002) reports high detection rates for pheochromocytomas when testing for plasma free
metanephrines, with sensitivities of “97% for familial and 100% for sporadic cases, [and] specificities of 96% for
familial and 80% for sporadic cases” (p 5). [Pacak (2002) provides a table comparing sensitivities and specificities
of various plasma and urine tests.] Kudva et al (2003) point out that whereas fractionated free plasma
metanephrines are highly sensitive in detecting hereditary pheochromocytomas, they have a high false positive
rate of detection for sporadic cases. They warn that “Even if the prevalence of pheochromocytomas was
estimated to be as high as one in every 200 patients screened, measurement of fractionated plasma metanephrines
would result in 30 patients with false positive tests for every one patient with pheochromocytoma detected” (p 8).
Pheochromocytomas secreting dopamine likely present with increased levels of plasma and urine DOPA and
dopamine. This is not a standard part of testing for pheochromocytomas and another reason to suspect that the
psychosis/schizophrenia connection with pheochromocytomas is possibly being missed. Pacak (2002) says that
“Elevations in plasma or urinary DOPA and dopamine are not in themselves particularly sensitive or specific
markers of benign or metastatic pheochromocytoma. However, when accompanied by elevations in plasma
norepinephrine or other clinical evidence [not that we necessarily recognize all the evidence, by any means] of
pheochromocytoma, such elevations should arouse immediate suspicion of metastatic disease” (p 6).
A recent report by Guller et al (2006) of testing 152 patients with pheochromocytomas (29.6% malignant, 23.0%
hereditary, 51.4% having spells, 66.6% with hypertension) found the following: Total urinary normetanephrine had
the highest sensitivity, at 96.9%; platelet norephinephrine 93.8%, and I-MIGB scintigraphy 83.7%. I-MIBG combined
with platelet norepinephrine had a sensitivity of 100%, with plasma norepinephrine 97.1%, with total urine
normetanephrine 96.6%, and with urine norepinephrine 95.3%. They advocate doing an MIBG scan if catechol-
amine levels are normal and a pheochromocytoma is suspected.
CT may identify 95%, MRIs perhaps more than 95%. CT and MRI specificity is only about 65-75%, however,
per Pacak,12 most commonly due to masses (often adrenal) that are not pheochromocytomas. MIBG is only
78-83% sensitive and is 95-100% specific12. Pacak12 says that 123I-MIBG is superior to 131I-MIBG. He further
indicates that 6-[18F] fluorodopamine positron emission tomography (PET) imaging shows promise as a tool
superior to MIBG for detecting and localizing pheochromocytomas and extra-adrenal tumors.
Pacak (2002) suggests doing abdominal CT or MRI if biochemical testing confirms pheochromocytoma.
If abdominal CT or MRI is negative, do whole-body CT or MRI. If CT or MRI is positive, abdominal or whole body, do
MIBG scan to determine whether the mass is a pheochromocytoma (remember <85% sensitive, however).
If CT, MRI, and MIBG are equivocal in the case of a positive biochemical test, Pacak recommends 6-[18F]fluoro-
dopamine PET scanning.
Ultrasound (U/S) imaging may be used as the initial imaging in certain patients such as infants and children
and pregnant women, per Ilias and Pacak (2004). However, they say U/S imaging is not better than MRI, and that
its specificity is thought to be low. Tumors in the neck, such as paragangliomas (head and neck paragangliomas
are also referred to as glomus tumors), can be incidentally detected in the neck by carotid U/S imaging done
widely to detect atherosclerotic plaque. The low incidence of carotid paragangliomas could be due in part to
mistaking tumors for plaque and under-detecting small tumors that could possibly be exerting a significant effect.
In the 1980s pheochromocytomas were identified about twice as often in the right adrenal gland as in the left
(Benowitz, 1988). [The right adrenal gland has a rather distinct pyramid shape, unlike the left side, incidentally.]
Slawik and Reincke (2003) point out that U/S and earlier imaging detected right-sided tumors more easily than
left-sided ones. They report that more recent studies indicate that pheochromocytomas may affect both adrenals
equally and that 2-10% may affect both adrenals simultaneously.
Many tumors are detected as incidentalomas, meaning, by definition, that they were not suspected until
discovered by an imaging procedure. That is not surprising, given the symptom variability of tumors and the lack of
suspicion with symptoms less commonly thought of as related to a tumor. I think we need to be more suspicious of
tumors and test for them much more frequently. We need to keep in mind that just because a tumor is an
incidentaloma, that does not mean that it is not significant. Slawik and Reincke (2003) report that “up to 20% of all
incidentalomas are significantly hormonally active” (p 7).
Testing for pheochromocytomas is obviously rather complex and we know that many have been missed.
Although research indicates that greater than 99% are missed, given the complexity of accurate testing, the
number of actual misdiagnosed cases could be far greater than many might imagine.
Cost of testing
Sawka et al (2004) estimated the costs of testing for pheochromocytomas, based on finding the tumors in
500 out of 100,000 patients with hypertension, using three different hypothetical algorithms, followed by abdominal
CT imaging. Algorithm A measured fractionated plasma metanephrines; algorithm B used 24-hour urine total
metanephrines or catecholamines; algorithm C used imaging only if fractionated plasma metanephrine was
> 0.5 nmol/liter or normetanephrine was > 1.80 nmol/liter; or for subjects in algorithm C with values between the
reference range and the above values, imaging would be done only if 24-hour urinary total metanephrines or
catecholamines were positive. They determined that algorithm A would cost 56.6 million dollars ($566 per person),
algorithm B 39.5 million dollars ($395 per person), and algorithm C 28.6 million dollars ($286 per person, on average).
Pheochromocytomas would be detected in an estimated 489, 457, and 478 of 500 persons with the tumors,
respectively, using algorithms A, B, and C.
Sawka et al (2004) indicate that if the approximately 12 million persons in the United States with uncontrolled
hypertension were screened using algorithm C that the cost would be an estimated 3.4 billion dollars. They point
out that this is about one-third of the yearly cost of hypertension in the United States. They suggest ways costs
can be reduced, involving improved biochemical testing, improved imaging techniques, and lower fees.
Widespread increases in testing could greatly drive down the cost per unit of testing. For example, if you buy a
million-dollar imaging machine and have to charge $1000 for each of 1000 patients to pay for the machine, you
could charge half as much for twice as many patients.
Following diagnosis of a pheochromocytoma, there is the cost of treatment, which often involves expensive
surgery. At best, testing the general population that has disease possibly related to pheochromocytomas, and
then treating those who have the tumors, would be an expensive proposal. Yet, detection and treatment of
disease-causing tumors could greatly reduce human suffering almost immediately and reduce other healthcare
costs over the long term. It might be worth it, actually, both in terms of good healthcare and cost. Good health is
worth every possible consideration.
Treatment [Please refer to updated treatment guidelines elsewhere, such as in the annual publication
Current Medical Diagnosis and Treatment, etc.]
The pharmacological treatment of choice for pheochromocytomas is alpha-adrenergic receptor blockers
until surgery or for metastatic or inoperable tumors. Beta blockers or combined alpha/beta blockers are used
to control arrhythmias such as tachycardia. Removal of the tumors by laparoscopic surgery is the ultimate
treatment of choice. Precautionary measures are essential to avoid life-threatening complications during
surgical removal of pheochromocytomas.
Conclusion
The statistics that 1% of persons has schizophrenia and 1% of 30,000 persons from 1944-1994 was found
at autopsy or by imaging to have adrenal tumors (not known how many were pheochromocytomas or other
active tumors) are interesting findings in reviewing the literature for this paper. This alone certainly does not
suggest a correlation. We do not know whether any of the persons with pheochromocytomas had schizophrenia.
However, we do know that pheochromocytomas can cause psychosis and that they (perhaps not all of them)
secrete dopamine, a catecholamine. We know that the dopamine hypothesis of schizophrenia prevails.
There is a multitude of other questions we need to answer about the effects of chromaffin cell tumors such as
pheochromocytomas and of other NETS. For example, are pheochromocytomas a major cause of anxiety?
Could the success of alpha blockers for BPH, really be due to their treating a misdiagnosed urinary bladder
pheochromocytoma in many cases? Could such problems as sleep apnea and sudden infant death be related to
a paraganglioma in the neck pressing on the carotid body, which exerts significant control over respiration and
cardiovascular responses? Does obesity (and smoking perhaps) lead to sleep apnea and hypoxia, which may
contribute to the development of carotid body tumors, as some studies suggest (Emedicine, 2005)? What are the
cardiac implications of tumors, whether or not they are in the heart or the neck or elsewhere, and whether or not
they are related to sleep apnea? How do NETS affect other bodily systems in ways we do not know or that we
know and under-diagnose? How significant a problem is it that we are missing many cases? Is there a cancer
connection we are missing? There are many additional extremely important questions that we need answered.
A greater than 99% diagnostic failure rate for pheochromocytomas for most of the twentieth century, which
the Mayo Clinic data (Sheps et al, 1990) suggest, is unacceptable in modern medicine. It could have vast
implications not only for conditions possibly related to pheochromocytomas but for conditions associated with
other NETS as well and for missing the causes of a multitude of disorders with differential diagnoses.
Schizophrenia and other psychotic disorders, anxiety, depression, etc could be just the tip of a very large
iceberg of disorders caused by non-diagnosed NETs. The toll in terms of human suffering that could be
alleviated could be tremendously greater than we realize.
References
Allolio B, 2001. Adrenal incidentalomas, in: Margioris CG (Eds). Adrenal Disorders, Totowa: Humana Press Inc.
Benabarre A, Bosch X, Plana MT, Lecube A, Vieta E, Cirera E, Valdes M, 2005. Relapsing paranoid psychosis as
the first manifestation of pheochromocytoma. J Clin Psychiatry. 66 (7) 949-950.
Benowitz NL, 1988. Pheochromocytoma—Recent advances in diagnosis and treatment [Medical Staff
Conference]. West J Med. 148 561-567.
Boorman III, Erie Vann, 2000. Pheochromocytoma masquerading for decades. Clinical Vignette. Proceedings
of UCLA Healthcare. 4 (2) 2-4.
Bravo EL, Tarazi RC, Gifford RW et al, 1979. Circulating and urinary catecholamines in pheochromocytoma.
N Engl J Med. 301 682-686.
Brown D, 2003. The Da Vinci Code. Anchor Books, New York.
Emedicine, June 10, 2005. Glomus Tumor (Head and Neck). Accessed at http://www.emedicine.com/radio/
topic309.htm.
Guller U, Turek J, Eubanks S, Delong ER, Oertli D, Feldman JM, 2006. Detecting pheochromocytoma: defining
the most sensitive test. Ann Surg. 243 (1) 102-107.
Harvard Consumer Health Information. Disease & Conditions. Mental Health, March 13, 2006. Schizophrenia.
Accessed at www.intelihealth.com/IH/ ihtIH/WSIHW000/8271/8694/188010.html?d=dmtHealthAZ.
Ilani T, Ben-Shachar D, Strous RD, Mazor M, Sheinkman A, Kotler M, Fuchs S, 2001. A peripheral marker for
schizophrenia: Increased levels of D3 dopamine receptor mRNA in blood lymphocytes. PNAS (Proceedings
of the National Academy of Sciences). 98 (2) 625-628.
Illias I, Pacak K, 2004. Current approaches and recommended algorithm for the diagnostic localization of
pheochromocytoma. J Clin Endocrinology & Metab. 89 (2) 479-491.
Jacomet P, 2001. Uncharted territory. VHL Family Alliance. Accessed at www.vhl.org/newsletter/vhl2001/
01adpheo.htm.
Kaltsas GA, Besser GM, Grossman AB, 2004. The diagnosis and medical management of advanced
neuroendocrine tumors. Endocrine Reviews. 25 (3) 458-511.
Kudva YC, Sawka AM, Young Jr WF, 2003. The laboratory diagnosis of adrenal pheochromocytoma:
The Mayo Clinic Experience. J of Clin Endocrinology & Metabolism. 88 (10) 4533-4539.
Kwak YT, Koo M-S, Choi, C-H, Sunwoo IN, 2001. Change of dopamine receptor mRNA expression in
lymphocyte of schizophrenic patients. BMC Med Genet. 2 (3) 1-10.
NCIPDQ Treatment–Health Professional Information, April 2006. Pheochromocytoma: Accessed at http://
health.yahoo.com/ency/healthwise/_ncicdr0000062928.
Pacak, K, March 1, 2002. Pheochromocytoma. Chapter 34. Endotext.com: Your Endocrine Source.
Accessed at http://www.endotext.org/adrenal/adrenal34/adrenal34.htm.
Proye C, Fossati P, Fontaine P, et al, 1986. Dopamine-secreting pheochromocytoma: an unrecognized entity?—
Classification of pheochromocytomas according to their type of secretion. Surgery. 100 1154-1162.
Sawka AM, Gafni A, Thabane L, Young Jr WF, 2004. The economic implications of three biochemical screening
algorithms for pheochromocytoma. J Clin Endocrin & Metab. 89 (6) 2859-2866.
Sheps SG, Jiang NS, Klee GG, van Heerden JA, 1990. Recent developments in the diagnosis and treatment of
pheochromocytoma. Mayo Clin Proc. 65 (1) 88-95.
Slawik M, Reincke M, Jan 22, 2003. Adrenal Incidentalomas. Chapter 20. Endotext.com. Accessed at http://
www.endotext.org/adrenal/adrenal20/adrenal20.htm.
Sutton MG, Sheps SG, Lie JT, 1981. Prevalence of clinically unsuspected pheochromocytoma. Review of a
50-year autopsy series. Mayo Clin Proc. 56 (6) 354-360.
Tierney Jr LM, McPhee SJ, Papadakis MA, (Eds), 2000. Current Medical Diagnosis and Treatment. 39th ed.
Lange Medical Books/McGraw-Hill, New York.
Walker E, Lewine RJ, 1990. Prediction of adult-onset schizophrenia from childhood home movies of the patients.
Am J Psychiatry. 147 (8) 1052-1056.
A subhypothesis of my NETS (neuroendocrine tumors) hypothesis of disease
[NETS and behavioral, mental, and physical disorders]
For Neuroendocrine tumor/Pheochromocytoma
website www.cherylswoffordnets.com
Author: Cheryl D. Swofford, MD,PsyD,MSci,MArts
Atlanta, Georgia and Marion, NC, USA
Last update: January 26, 2007
Word counts: Abstract: 175 Text: 4000
Keywords: neuroendocrine tumors, NETS, pheochromocytoma, schizophrenia, hypertension,
anxiety, paraganglioma, psychosis, adrenal tumor, chromaffin cell tumor
“The miniature structure itself protrudes up through the floor as though it were the tip of an iceberg—
the apex of an enormous, pyramidical vault, submerged below like a hidden chamber.”
Dan Brown. In The Da Vinci Code (p. 488).
I hypothesize that pheochromocytomas (and paragangliomas) are a non-diagnosed cause of many cases of
psychiatric disorders, in particular anxiety, schizophrenia and other psychotic disorders. Further, I hypothesize
that neuroendocrine tumors (NETs) (of which pheochromocytomas and paragangliomas are two of many
different types) are a non-diagnosed cause of a significant number of cases of mental and physical conditions
[such as, asthma, benign prostatic hypertrophy or hyperplasia (BPH), blindness, cancers, cardiac disorders,
cranial neuropathy, headache, seizures, and other neurological disorders, depression, diabetes, hearing loss,
hypoglycemia, hyperthyroidism and other endocrine disorders, eclampsia/pre-eclampsia of pregnancy,
homosexuality, hyperactivity, hypertension, indigestion, menopausal symptoms, Raynaud’s, sudden infant death
syndrome, sleep apnea and other sleep disorders, etc.] However, NETs other than pheochromocytomas and
paragangliomas, as well as non-psychiatric disorders, are beyond the scope of this paper. I plan to address
those topics in future papers and a book.
Pheochromocytomas and paragangliomas are the two subtypes of chromaffin cell tumors. Pheochromo-
cytomas come from the adrenal medulla, and paragangliomas are extra-adrenal, derived from sympathetic or
parasympathetic paraganglia. A widely-recognized “rule of 10” roughly applies to chromaffin cell tumors:
10% extra-adrenal, 10% malignant, 10% familial, 10% not associated with hypertension, 10% bilateral, and 10%
in children (Kaltsas, Besser & Grossman, 2004). The remainder of this paper will refer to pheochromocytomas
primarily, since they are thought to account for about 90% of these tumors. [However, keep in mind that these
tumors are vastly under-diagnosed and we really do not know the numbers.] [Paragangliomas have been found
in the heart, along the para-aortic sympathetic chain, at the organ of Zuckerkandl (at the aortic bifurcation), in the
neck (carotid body tumor or glomus tumor), the mediastinum, and in and around the urinary bladder. They can
exist anywhere there are chromaffin cells. Less than 1% of chromaffin cell tumors have been found to be extra-
abdominal (Benowitz, 1988).
Pheochromocytomas are known for their sympathetic nervous system (SNS) manifestations, due to their
release of neurotransmitters, the catecholamines epinephrine (or adrenaline) and norepinephrine (or noradren-
aline). The tumors also release the catecholamine dopamine in lesser amounts and fewer cases, it is thought.
I hypothesize that any bodily function affected by the SNS and catecholamines has the potential of being affected
by pheochromocytomas. The primary focus of this article will be schizophrenia and its possible connection with
dopamine, with the underlying assumption that psychotic disorders in general may be related to pheochromo-
cytomas. Anxiety and other disorders will also be mentioned.
Types
Pheochromocytomas may be genetic or sporadic, local and “benign,” non-malignant or malignant, non-
metastatic or metastatic. [To say that they are sporadic to suggest they are not genetic may not be accurate.
To say they are “benign” because they are not malignant is more or less a misnomer.] They may go undiagnosed
and cause few to many problems. They may be diagnosed and an association may not be made with all the
problems they have caused. They may be diagnosed and cause variable problems in an unpredictable, varying
manner.
Causes
The causes of most pheochromocytomas, other than perhaps metastatic tumors, are unknown. Most are
thought to be caused by genetic mutations.3 An estimated 10-20% of identified pheochromocytomas are found
in patients with four well-known disorders due to gene mutations causing an autosomal dominant pattern of
inheritance: multiple endocrine neoplasia syndromes (MEN 2A and 2B), von Hippel-Lindau disease (VHL) and
neurofibromatosis type 1 (NF1).
Slawik and Reincke (2003) report that “Adrenal nodularity increases with age and usually affects both glands”
(p. 17), leading to the hypothesis of aging as a cause of adrenal tumors. They explain, however, that most of the
tumors have a monoclonal origin, the result of an oncogenic mutation and that a lesser percentage of cases result
from polyclonal nodular hyperplasia in aging. They estimate that 75% of nonadrenal malignant tumors metastasize
to the adrenal glands (25-100%, depending on the type of cancer). This suggests a highly significant interaction
with the adrenals and other bodily systems and, possibly, with pheochromocytomas as well.
Although pheochromocytomas may not be a disease of aging, necessarily, age can be a factor. We are more
prone to heart disease, cancers and other tumors, etc as we age. For example, some paragangliomas, such as
carotid body tumors, with possible links to sleep apnea, and respiratory and cardiac problems, have been found to
have a peak incidence at ages 45-50 (Emedicine, 2005). [Since many persons in our society also start having
problems around that age thought to be due to aging and genetics and lifestyle, such as hypertension and heart
disease, it may be unlikely that a tumor is suspected, when it could be causing serious problems.]
Clinical manifestations
Clinical manifestations of diagnosed pheochromocytomas are primarily severe headache, palpitations,
profuse sweating, and hypertension. All diagnosed (with pheochromocytoma) patients have not had all of these
signs or symptoms.
Pheochromocytomas can also cause psychosis, paranoia, anxiety, and confusion; they do not in every case,
or even in most cases, as far as is known. But I hypothesize that pheochromocytomas may be a major cause
and at least a more significant cause of the aforementioned disorders than is now recognized. Much is still not
known about the cases they cause.
In our age of political correctness, and with the stigma attached to certain disorders over many years, it is
highly possible that studies that examined pheochromocytomas failed to report whether any of the patients had
a psychotic disorder or some other psychiatric disorder. [I have seen many charts of patients diagnosed with
schizophrenia or HIV, for example, with non-psychiatry or non-infectious disease physician notes that made little
or no mention of these diagnoses, even when it could have been relevant to the case.] Patients presenting with
psychiatric disorders by psychiatrists and non-psychiatrists generally are not tested for pheochromocytomas.
Non-psychiatric physicians generally are not looking for the cause of schizophrenia at all.
Pheochromocytomas have also been linked with causing vasomotor changes, facial pallor, bradycardia,
tachycardia, precordial pain, abdominal pain, anxiety, nervousness, irritability, weight loss, increased appetite,
angina, hypertension, cardiac enlargement, postural hypotension, elevated temperature, retinal and cerebro-
vascular hemorrhage, sudden blindness, seizures, hyperglycemia, paresthesias, Raynaud’s, pulmonary edema,
heart failure, cardiomyopathy, Cushing’s syndrome, erythrocytosis, hypercalcemia, hypermetabolism (with
normal thyroid function tests), thyrotoxicosis, hyperglycemia, etc (Tierney et al, eds., 2000). Invasive diagnostic
procedures and/or surgery in patients with pheochromocytomas that have not been diagnosed, can lead to severe
hypotension and shock, renal failure, and myocardial infarction (Tierney et al, eds, 2000).
According to Benowitz (1988) “Abnormal carbohydrate metabolism, particularly the sudden, unexpected
development of diabetes mellitus, should suggest a pheochromocytoma” (p. 564). Also, “Unusual hypertensive
responses to anesthesia, to any antihypertensive drugs (particularly beta-blockers), or to sympathomimetic
drugs are useful clues to the presence of the tumor” (p. 564). Patients in a hypercalcemic state may also exhibit
symptoms caused by calcium as a trigger for the release of catecholamines by a pheochromocytoma (Benowitz,
1988).
There is no reason to think that many of these signs or symptoms have a greater incidence in a schizo-
phrenia patient population than in the general patient population. However, it is known that at least one of the
above symptoms, psychosis, has a greater incidence among persons with schizophrenia. So why not examine
pheochromocytomas in more detail in schizophrenia patients? We need to keep in mind, however, that the use
of additional medications among schizophrenia patients vastly complicates getting a clear picture of possible
pheochromocytoma effects. There is likely a huge number of patients diagnosed with schizophrenia today,
started on antipsychotic drugs, and not adequately examined and tested for pheochromocytomas or not tested
prior to starting new drugs that could influence test results.
Under-diagnosing
Pheochromocytomas may be involved in any of the preceding problems but not diagnosed. A 1990
(Sheps, Jiang et al) report of data from Mayo Clinic indicates that only 1-2 of 300 out of 100,000 persons with
pheochromocytomas at autopsy are diagnosed in life (0.3% incidence of pheochromocytomas in autopsy cases,
with more than 99% non-diagnosed). [Mayo Clinic started keeping data on adrenal tumors when Dr. C. H. Mayo
removed such a tumor in 1926 (Sheps et al, 1990).] Allolio (2001) found a 1% prevalence of adrenal tumors in
more than 30,000 patients in autopsy series and computed tomography (CT) studies from 1944-1994. This
finding strongly suggests the need for imaging such as CT to detect the tumors.
Pheochromocytomas have been widely missed and “misdiagnosed as…anxiety, tension states, psycho-
neurosis, psychosis, hypoglycemia, menopause, headaches [including migraines and other types],…drug abuse:
amphetamines, cocaine, LSD, phenylpropanolamine, caffeine,…interaction of monoamine oxidase inhibitors with
certain food and beverages,…clonidine withdrawal,…atropine administration, [use of] nasal decongestants [and]
tricyclic antidepressants,…trauma, tumor, stroke, focal arterial insufficiency, autonomic hyperreflexia,…seizure,
…angina pectoris, exaggerated response to hypovolemia or infection,…tachycardia, familial dysautonomia
(Jacomet, 2001). Interestingly, a recent paper reported a case in which pheochromocytoma was found where
relapsing paranoid psychosis was described as the first manifestation of the tumor (Benabarre et al, 2005).
Are we onto something of much greater significance here, not yet revealed because of all the currently missed
cases?
Pheochromocytomas are considered rare, and statistics are largely limited to co-occurrence with hyper-
tension, occurring in an estimated one to five per 1000 cases of hypertension (NCIPDQ 2006). Given that an
estimated 20% of Americans have hypertension, and an estimated 1% of Americans [and persons worldwide]
(Harvard Consumer Health Information, 2006) has schizophrenia, we should be looking more closely at
pheochromocytomas as a cause, starting with all schizophrenia patients with uncontrolled hypertension and
expanding our search as much as possible. Ideally, we should not limit testing to hypertensives, however, given
the symptom variability, mentioned later.
Prior to starting this paper, I was in a position to partially examine each of more than 10,000 cardiology
patient charts, many for patients with hypertension and/or other cardiovascular problems. I read the most
recent assessment and plan in all of these patient charts. However, I recall fewer than five cases in which
testing for pheochromocytoma was mentioned. I worked in a department of psychiatry at a major university
teaching hospital for several years, working with schizophrenia researchers. I did not have pheochromo-
cytomas in mind and was not doing chart reviews related to pheochromocytomas. I do not recall anyone
mention pheochromocytomas as a possible cause of psychiatric disorders. The physicians and other
clinicians and researchers were experts in their fields, competent and respected. This is not to suggest
that they were not doing their jobs by current standards. I am suggesting that further testing for
pheochromocytomas, especially among hypertensive and schizophrenia patient populations, could reveal
more insight into the role of pheochromocytomas in hypertension as well as schizophrenia and other mental
and physical disorders.
Dopamine
An estimated 10% of pheochromocytomas are malignant. Malignant pheochromocytomas appear more
likely to produce and secrete more dopamine. According to Pacak (2002) elevated plasma or urinary dopamine,
“when accompanied by elevations in plasma norepinephrine or other clinical evidence of pheochromocytoma…
should arouse immediate suspicion of metastatic disease” (p. 6). An estimated 10-15% of persons with
schizophrenia die by suicide. The factors involved in the suicide of a person with schizophrenia are complex,
needless to say. Is it possible, though, that among the most ill patients with schizophrenia are those with
malignant, dopamine-secreting pheochromcytomas, and they die by suicide without being diagnosed? Since
we do not have the answers to these questions, we need to be seeking them. We now have the means.
It was not until the 1980s that dopamine release by pheochromocytomas became recognized in medical and
psychiatry circles. As recently as 1986, Proye et al (1986) published a paper questioning dopamine as an
unrecognized entity in pheochromocytomas. Yet, the possibility of pheochromocytomas as a significant cause
of psychotic illnesses has not been explored, although the dopamine hypothesis of psychotic illness continues.
The dopamine hypothesis of schizophrenia never goes away. Early work related to this hypothesis came
about when it was found that drugs that block brain dopamine function reduced psychotic symptoms.
[Near the mid-twentieth century chlorpromazine was used to control blood pressure elevated by certain
anesthesias used during surgeries. Then people began to notice that chlorpromazine reduced psychosis in
psychotic surgery patients; thus the advent of antipsychotic drugs. The connection with dopamine, hypertension,
and pheochromocytomas will not go away, either. Many anti-psychotic drugs can cause both hypotension and
neuroleptic malignant syndrome, with malignant hypertension. This subject is not simple.]
The brain has been studied in nearly every way imaginable. Still, the cause of schizophrenia and other
psychotic disorders is a mystery. However, evidence of excess brain dopamine remains. My hypothesis is that
we need to look outside the brain for the source of excess dopamine, to wit, pheochromocytomas and possibly
other NETs.
There is strong evidence of a correlation between peripheral dopamine and schizophrenia.
Ilani, Ben-Shachar et al (2001), with Professor Sara Fuchs, say that “Although the precise pathophysiology of
schizophrenia remains unknown, the dopaminergic hypothesis of the illness assumes that the illness results from
excessive activity at dopamine synapses in the brain” (p 625). They looked beyond the brain. They found
evidence of increased levels of peripheral dopamine (the D3 subtype) in schizophrenia patients, with
“a significant elevation of at least 2-fold in the mRNA level of the D3, but not the D4, dopamine receptor in
schizophrenic patients” (p. 625). They specifically looked at dopamine receptors on peripheral blood
lymphocytes, as did Kwak, Koo et al (2001) around the same time.
Pheochromocytomas have been found at autopsy in babies from two days old to adults 92 years old
(Sutton et al, 1981). Psychotic disorders have been diagnosed in children as well as adults (so has hypertension).
A study of home movies of children indicated that viewers accurately identified those children who later developed
schizophrenia (Walker & Lewine, 1990). Pheochromocytomas can enlarge with age, possibly growing larger
earlier in males, perhaps in a pattern consistent with schizophrenia. Much is going on that has not been explained
adequately. Let’s look at pheochromocytomas.
We know that pheochromocytomas can “masquerad[e] for decades” (Boorman & Erie Vann, 2000, p 2), as in
the case of a 38-year-old man who, in retrospect, had bizarre, hypervigilant, hyperactive behavior in high school.
He was found to have an elevated level of norepinephrine when diagnosed with a “benign” 3 cm pheochromocytoma
shown by an MRI. “His world had been veiled by the misperceptions caused by the pheochromocytoma, but the
veil was removed” (p 2). How many children may be hyperactive and how many adults may have anxiety because
they have a pheochromocytoma that no one knows about? We have no idea. There may be other tumors producing
dopamine and causing psychosis, or epinephrine or other chemicals causing anxiety and other disorders that no
one knows about. How many people are out there suffering, with tumors secretly secreting chemicals that are
negatively affecting them in all sorts of ways, behaviorally, mentally, and physically? We have the tools and we
could find out.
Symptom variability
According to Benowitz (1988) “the pattern of catecholamine release and related clinical manifestations [are]
variable” (p 561) in pheochromocytomas. The common symptoms, as we know them, of pheochromocytomas,
spells of severe headache, sweating, and palpitations, occur in about 90% of diagnosed patients. [One can see
that such patients might be misdiagnosed with anxiety, or have anxiety due to a pheochromocytoma.] But what are
the symptoms in the estimated greater than 99% of patients who are not diagnosed? Patients do not necessarily
exhibit all three of the most commonly known symptoms, but having these three symptoms together is considered a
strong diagnostic clue. There can be opposite symptoms and disagreement about symptoms. For example, pallor is
considered common, but some books will say flushing is a common symptom; others say flushing is rare
(Benowitz, 1988). Bradycardia is a common reflex response to norepinephrine-related vasoconstriction. But
tachycardia is perhaps a less common effect of epinephrine released by these tumors.
Hypertension, hypermetabolism, and weight loss are other commonly known manifestations. Dopamine and
epinephrine-producing tumors have, however, been found with a hypermetabolic state and normal blood pressures
(Benowitz, 1988). More than 90% of persons diagnosed with pheochromocytomas have hypertension (Benowitz,
1988). However, fifty years of Mayo Clinic data found hypertension in only 54% of patients diagnosed with
pheochromocytoma at autopsy (Sutton et al, 1981). I hypothesize that pheochromocytomas are mainly diagnosed
only in persons with hypertension and largely non-diagnosed in those who do not have hypertension or have
hypertension that is considered idiopathic, “essential” hypertension without testing for the tumors. Since
hypertension and the most commonly known manifestations of pheochromocytomas are not the only signs and
symptoms, we need to be looking far beyond those patients for pheochromocytomas.
Bravo et al (1979) did not find a strong correlation with circulating catecholamine levels and blood pressure in
pheochromocytoma, as one might expect. I hypothesize that the lack of correlation may also apply to other clinical
manifestations of pheochromocytomas. Such symptom variability would reduce suspicion of pheochromocytomas
and suggests a strong likelihood of under-diagnosing the tumors, perhaps a much greater likelihood than we might
suspect. Benowitz (1988), citing a variety of research, explains some of the reasons for such variability, related to
enzyme activity, hypovolemia, the release of both vasoconstrictor and vasodilator substances by the tumors, down-
regulation, sympathetic neural control of blood pressure, and such factors as change in posture, exercise, and
emotional state. The physiology is quite complex, causing confusion, misunderstanding, and vast under-diagnosing.
However, we have the diagnostic tools for rather accurate detection if we have the wherewithal.
Testing
Diagnostic testing modalities for pheochromocytomas have advanced greatly in recent years. Physicians/
scientists are still under-testing for pheochromocytomas and are still missing a large of percentage of cases.
Physicians are misdiagnosing a great number of cases of diseases heretofore not recognized as being related to
pheochromocytomas and that can be treated by reducing or removing tumors. Pheochromocytomas are still
considered rare, but they may be much more common than we know. Testing for them, and adequate testing, may
be what is rare. One physician recently told me, “Pheos are a pain to diagnose, and you have to keep testing and
testing.”
When testing is performed it often involves collecting 24-hour urinary vanillylmandelic acid (VMA),
catecholamines and metanephrines. [Catechol-o-methyltransferase (COMT) and monoamine oxidase (MAO) are
the two major enzymes that metabolize catecholamines to produce VMA; metanephrines are break-down products
of epinephrine.] Such urine testing can result in a “false negative” if the person being tested did not have a recent
exacerbation of symptoms related to catecholamine release or did not follow instructions. Instructions are to avoid
alcohol, coffee, tea, tobacco, and strenuous exercise for three days prior to testing. The same applies to blood
testing, plus at least eight hours of fasting (with only water), and either lying supine for 30 minutes or sitting upright
for 15 minutes are required.
Antipsychotic and antidepressant drugs and any drugs that influence the sympathetic nervous system can
affect results. Benowitz (1988) points out that calcium channel blockers can decrease “urinary catecholamine
secretion in patients with pheochromocytoma” (p. 566). Pacak (2002) recommends stopping acetaminophen for
at least 5 days prior to blood testing for pheochromocytomas. Pacak further points out that impaired renal function
(as in patients with congestive heart failure and hypertension) can cause high values for metabolic end-products
that can be misinterpreted as a pheochromocytoma.
Benowitz (1988, p 564) says that “Assays of urinary catecholamines and metanephrines are… adequate to
diagnose pheochromocytoma in most patients, but when the results of these studies are equivocal and
pheochromocytoma is suggested clinically, plasma catecholamines should be measured.” “There may be
considerable swings in the levels of plasma catecholamines from time to time, a consequence of postural changes,
exercise, emotional arousal, and other factors. Despite such variability, basal concentrations are usually increased
severalfold in pts with pheochromocytoma.” “To diagnose a dopamine-secreting pheochromocytoma, assays of
urinary dopamine may be much more sensitive.”
Imaging techniques used to locate pheochromocytomas are computed tomography (CT) scan, magnetic
resonance imaging (MRI), and metaiodobenzylguanidine (MIBG) scan. MRI has the advantages of avoiding IV
contrast material and irradiation and better differentiating tissue than CT. MIBG (or I-MIBG), using radioactive iodine
(usually 131I in the United States) is more sensitive and specific and can detect better than CT or MRI whether a
mass is a functioning pheochromocytoma or a benign adenoma.
Urine testing is generally considered more useful than blood testing for pheochromocytomas. That is not
necessarily true; it appears to depend on the type of test and type of pheochromocytoma, whether it is genetic or
sporadic, the age of the patients, and what the tumor is secreting when. Testing of 24-hour urine provides
“acceptable sensitivity and significantly better specificity than fractionated plasma free metanephrine,” per
Kudva et al (2003) in cases of sporadic pheochromocytoma presenting with triggers such as resistant hyper-
tension and spells (headache, profuse sweating, and tachycardia). However, Pacak (2002) recommends
fractionated plasma free metanephrine testing in the young pediatrics population for practical purposes.
Pacak (2002) reports high detection rates for pheochromocytomas when testing for plasma free
metanephrines, with sensitivities of “97% for familial and 100% for sporadic cases, [and] specificities of 96% for
familial and 80% for sporadic cases” (p 5). [Pacak (2002) provides a table comparing sensitivities and specificities
of various plasma and urine tests.] Kudva et al (2003) point out that whereas fractionated free plasma
metanephrines are highly sensitive in detecting hereditary pheochromocytomas, they have a high false positive
rate of detection for sporadic cases. They warn that “Even if the prevalence of pheochromocytomas was
estimated to be as high as one in every 200 patients screened, measurement of fractionated plasma metanephrines
would result in 30 patients with false positive tests for every one patient with pheochromocytoma detected” (p 8).
Pheochromocytomas secreting dopamine likely present with increased levels of plasma and urine DOPA and
dopamine. This is not a standard part of testing for pheochromocytomas and another reason to suspect that the
psychosis/schizophrenia connection with pheochromocytomas is possibly being missed. Pacak (2002) says that
“Elevations in plasma or urinary DOPA and dopamine are not in themselves particularly sensitive or specific
markers of benign or metastatic pheochromocytoma. However, when accompanied by elevations in plasma
norepinephrine or other clinical evidence [not that we necessarily recognize all the evidence, by any means] of
pheochromocytoma, such elevations should arouse immediate suspicion of metastatic disease” (p 6).
A recent report by Guller et al (2006) of testing 152 patients with pheochromocytomas (29.6% malignant, 23.0%
hereditary, 51.4% having spells, 66.6% with hypertension) found the following: Total urinary normetanephrine had
the highest sensitivity, at 96.9%; platelet norephinephrine 93.8%, and I-MIGB scintigraphy 83.7%. I-MIBG combined
with platelet norepinephrine had a sensitivity of 100%, with plasma norepinephrine 97.1%, with total urine
normetanephrine 96.6%, and with urine norepinephrine 95.3%. They advocate doing an MIBG scan if catechol-
amine levels are normal and a pheochromocytoma is suspected.
CT may identify 95%, MRIs perhaps more than 95%. CT and MRI specificity is only about 65-75%, however,
per Pacak,12 most commonly due to masses (often adrenal) that are not pheochromocytomas. MIBG is only
78-83% sensitive and is 95-100% specific12. Pacak12 says that 123I-MIBG is superior to 131I-MIBG. He further
indicates that 6-[18F] fluorodopamine positron emission tomography (PET) imaging shows promise as a tool
superior to MIBG for detecting and localizing pheochromocytomas and extra-adrenal tumors.
Pacak (2002) suggests doing abdominal CT or MRI if biochemical testing confirms pheochromocytoma.
If abdominal CT or MRI is negative, do whole-body CT or MRI. If CT or MRI is positive, abdominal or whole body, do
MIBG scan to determine whether the mass is a pheochromocytoma (remember <85% sensitive, however).
If CT, MRI, and MIBG are equivocal in the case of a positive biochemical test, Pacak recommends 6-[18F]fluoro-
dopamine PET scanning.
Ultrasound (U/S) imaging may be used as the initial imaging in certain patients such as infants and children
and pregnant women, per Ilias and Pacak (2004). However, they say U/S imaging is not better than MRI, and that
its specificity is thought to be low. Tumors in the neck, such as paragangliomas (head and neck paragangliomas
are also referred to as glomus tumors), can be incidentally detected in the neck by carotid U/S imaging done
widely to detect atherosclerotic plaque. The low incidence of carotid paragangliomas could be due in part to
mistaking tumors for plaque and under-detecting small tumors that could possibly be exerting a significant effect.
In the 1980s pheochromocytomas were identified about twice as often in the right adrenal gland as in the left
(Benowitz, 1988). [The right adrenal gland has a rather distinct pyramid shape, unlike the left side, incidentally.]
Slawik and Reincke (2003) point out that U/S and earlier imaging detected right-sided tumors more easily than
left-sided ones. They report that more recent studies indicate that pheochromocytomas may affect both adrenals
equally and that 2-10% may affect both adrenals simultaneously.
Many tumors are detected as incidentalomas, meaning, by definition, that they were not suspected until
discovered by an imaging procedure. That is not surprising, given the symptom variability of tumors and the lack of
suspicion with symptoms less commonly thought of as related to a tumor. I think we need to be more suspicious of
tumors and test for them much more frequently. We need to keep in mind that just because a tumor is an
incidentaloma, that does not mean that it is not significant. Slawik and Reincke (2003) report that “up to 20% of all
incidentalomas are significantly hormonally active” (p 7).
Testing for pheochromocytomas is obviously rather complex and we know that many have been missed.
Although research indicates that greater than 99% are missed, given the complexity of accurate testing, the
number of actual misdiagnosed cases could be far greater than many might imagine.
Cost of testing
Sawka et al (2004) estimated the costs of testing for pheochromocytomas, based on finding the tumors in
500 out of 100,000 patients with hypertension, using three different hypothetical algorithms, followed by abdominal
CT imaging. Algorithm A measured fractionated plasma metanephrines; algorithm B used 24-hour urine total
metanephrines or catecholamines; algorithm C used imaging only if fractionated plasma metanephrine was
> 0.5 nmol/liter or normetanephrine was > 1.80 nmol/liter; or for subjects in algorithm C with values between the
reference range and the above values, imaging would be done only if 24-hour urinary total metanephrines or
catecholamines were positive. They determined that algorithm A would cost 56.6 million dollars ($566 per person),
algorithm B 39.5 million dollars ($395 per person), and algorithm C 28.6 million dollars ($286 per person, on average).
Pheochromocytomas would be detected in an estimated 489, 457, and 478 of 500 persons with the tumors,
respectively, using algorithms A, B, and C.
Sawka et al (2004) indicate that if the approximately 12 million persons in the United States with uncontrolled
hypertension were screened using algorithm C that the cost would be an estimated 3.4 billion dollars. They point
out that this is about one-third of the yearly cost of hypertension in the United States. They suggest ways costs
can be reduced, involving improved biochemical testing, improved imaging techniques, and lower fees.
Widespread increases in testing could greatly drive down the cost per unit of testing. For example, if you buy a
million-dollar imaging machine and have to charge $1000 for each of 1000 patients to pay for the machine, you
could charge half as much for twice as many patients.
Following diagnosis of a pheochromocytoma, there is the cost of treatment, which often involves expensive
surgery. At best, testing the general population that has disease possibly related to pheochromocytomas, and
then treating those who have the tumors, would be an expensive proposal. Yet, detection and treatment of
disease-causing tumors could greatly reduce human suffering almost immediately and reduce other healthcare
costs over the long term. It might be worth it, actually, both in terms of good healthcare and cost. Good health is
worth every possible consideration.
Treatment [Please refer to updated treatment guidelines elsewhere, such as in the annual publication
Current Medical Diagnosis and Treatment, etc.]
The pharmacological treatment of choice for pheochromocytomas is alpha-adrenergic receptor blockers
until surgery or for metastatic or inoperable tumors. Beta blockers or combined alpha/beta blockers are used
to control arrhythmias such as tachycardia. Removal of the tumors by laparoscopic surgery is the ultimate
treatment of choice. Precautionary measures are essential to avoid life-threatening complications during
surgical removal of pheochromocytomas.
Conclusion
The statistics that 1% of persons has schizophrenia and 1% of 30,000 persons from 1944-1994 was found
at autopsy or by imaging to have adrenal tumors (not known how many were pheochromocytomas or other
active tumors) are interesting findings in reviewing the literature for this paper. This alone certainly does not
suggest a correlation. We do not know whether any of the persons with pheochromocytomas had schizophrenia.
However, we do know that pheochromocytomas can cause psychosis and that they (perhaps not all of them)
secrete dopamine, a catecholamine. We know that the dopamine hypothesis of schizophrenia prevails.
There is a multitude of other questions we need to answer about the effects of chromaffin cell tumors such as
pheochromocytomas and of other NETS. For example, are pheochromocytomas a major cause of anxiety?
Could the success of alpha blockers for BPH, really be due to their treating a misdiagnosed urinary bladder
pheochromocytoma in many cases? Could such problems as sleep apnea and sudden infant death be related to
a paraganglioma in the neck pressing on the carotid body, which exerts significant control over respiration and
cardiovascular responses? Does obesity (and smoking perhaps) lead to sleep apnea and hypoxia, which may
contribute to the development of carotid body tumors, as some studies suggest (Emedicine, 2005)? What are the
cardiac implications of tumors, whether or not they are in the heart or the neck or elsewhere, and whether or not
they are related to sleep apnea? How do NETS affect other bodily systems in ways we do not know or that we
know and under-diagnose? How significant a problem is it that we are missing many cases? Is there a cancer
connection we are missing? There are many additional extremely important questions that we need answered.
A greater than 99% diagnostic failure rate for pheochromocytomas for most of the twentieth century, which
the Mayo Clinic data (Sheps et al, 1990) suggest, is unacceptable in modern medicine. It could have vast
implications not only for conditions possibly related to pheochromocytomas but for conditions associated with
other NETS as well and for missing the causes of a multitude of disorders with differential diagnoses.
Schizophrenia and other psychotic disorders, anxiety, depression, etc could be just the tip of a very large
iceberg of disorders caused by non-diagnosed NETs. The toll in terms of human suffering that could be
alleviated could be tremendously greater than we realize.
References
Allolio B, 2001. Adrenal incidentalomas, in: Margioris CG (Eds). Adrenal Disorders, Totowa: Humana Press Inc.
Benabarre A, Bosch X, Plana MT, Lecube A, Vieta E, Cirera E, Valdes M, 2005. Relapsing paranoid psychosis as
the first manifestation of pheochromocytoma. J Clin Psychiatry. 66 (7) 949-950.
Benowitz NL, 1988. Pheochromocytoma—Recent advances in diagnosis and treatment [Medical Staff
Conference]. West J Med. 148 561-567.
Boorman III, Erie Vann, 2000. Pheochromocytoma masquerading for decades. Clinical Vignette. Proceedings
of UCLA Healthcare. 4 (2) 2-4.
Bravo EL, Tarazi RC, Gifford RW et al, 1979. Circulating and urinary catecholamines in pheochromocytoma.
N Engl J Med. 301 682-686.
Brown D, 2003. The Da Vinci Code. Anchor Books, New York.
Emedicine, June 10, 2005. Glomus Tumor (Head and Neck). Accessed at http://www.emedicine.com/radio/
topic309.htm.
Guller U, Turek J, Eubanks S, Delong ER, Oertli D, Feldman JM, 2006. Detecting pheochromocytoma: defining
the most sensitive test. Ann Surg. 243 (1) 102-107.
Harvard Consumer Health Information. Disease & Conditions. Mental Health, March 13, 2006. Schizophrenia.
Accessed at www.intelihealth.com/IH/ ihtIH/WSIHW000/8271/8694/188010.html?d=dmtHealthAZ.
Ilani T, Ben-Shachar D, Strous RD, Mazor M, Sheinkman A, Kotler M, Fuchs S, 2001. A peripheral marker for
schizophrenia: Increased levels of D3 dopamine receptor mRNA in blood lymphocytes. PNAS (Proceedings
of the National Academy of Sciences). 98 (2) 625-628.
Illias I, Pacak K, 2004. Current approaches and recommended algorithm for the diagnostic localization of
pheochromocytoma. J Clin Endocrinology & Metab. 89 (2) 479-491.
Jacomet P, 2001. Uncharted territory. VHL Family Alliance. Accessed at www.vhl.org/newsletter/vhl2001/
01adpheo.htm.
Kaltsas GA, Besser GM, Grossman AB, 2004. The diagnosis and medical management of advanced
neuroendocrine tumors. Endocrine Reviews. 25 (3) 458-511.
Kudva YC, Sawka AM, Young Jr WF, 2003. The laboratory diagnosis of adrenal pheochromocytoma:
The Mayo Clinic Experience. J of Clin Endocrinology & Metabolism. 88 (10) 4533-4539.
Kwak YT, Koo M-S, Choi, C-H, Sunwoo IN, 2001. Change of dopamine receptor mRNA expression in
lymphocyte of schizophrenic patients. BMC Med Genet. 2 (3) 1-10.
NCIPDQ Treatment–Health Professional Information, April 2006. Pheochromocytoma: Accessed at http://
health.yahoo.com/ency/healthwise/_ncicdr0000062928.
Pacak, K, March 1, 2002. Pheochromocytoma. Chapter 34. Endotext.com: Your Endocrine Source.
Accessed at http://www.endotext.org/adrenal/adrenal34/adrenal34.htm.
Proye C, Fossati P, Fontaine P, et al, 1986. Dopamine-secreting pheochromocytoma: an unrecognized entity?—
Classification of pheochromocytomas according to their type of secretion. Surgery. 100 1154-1162.
Sawka AM, Gafni A, Thabane L, Young Jr WF, 2004. The economic implications of three biochemical screening
algorithms for pheochromocytoma. J Clin Endocrin & Metab. 89 (6) 2859-2866.
Sheps SG, Jiang NS, Klee GG, van Heerden JA, 1990. Recent developments in the diagnosis and treatment of
pheochromocytoma. Mayo Clin Proc. 65 (1) 88-95.
Slawik M, Reincke M, Jan 22, 2003. Adrenal Incidentalomas. Chapter 20. Endotext.com. Accessed at http://
www.endotext.org/adrenal/adrenal20/adrenal20.htm.
Sutton MG, Sheps SG, Lie JT, 1981. Prevalence of clinically unsuspected pheochromocytoma. Review of a
50-year autopsy series. Mayo Clin Proc. 56 (6) 354-360.
Tierney Jr LM, McPhee SJ, Papadakis MA, (Eds), 2000. Current Medical Diagnosis and Treatment. 39th ed.
Lange Medical Books/McGraw-Hill, New York.
Walker E, Lewine RJ, 1990. Prediction of adult-onset schizophrenia from childhood home movies of the patients.
Am J Psychiatry. 147 (8) 1052-1056.
CHERYL D. SWOFFORD MD, PsyD, MScience, MArts
RESEARCH on Neuroendocrine Tumors
(NETS)
RESEARCH on Neuroendocrine Tumors
(NETS)
