The case
At the age of 52 years, a man had begun to exhibit jaw tremor, decreased vocal volume, reduced facial expression and muscle pain, all of which suggested a diagnosis of Parkinson disease (PD). His symptoms improved on treatment with pramipexole and selegiline. Over the next 7 years, varying doses of carbidopa–levodopa–entacapone and controlled-release carbidopa–levodopa 50/200 were added to the drug regimen to control worsening PD symptoms. When the patient was 58 years old, he exhibited mild orthostatic hypotension during a clinic visit (supine blood pressure and heart rate 119/76 mmHg and 48 bpm, respectively; standing blood pressure and heart rate 103/71 mmHg and 53 bpm, respectively). This observation was considered to be secondary to PD or dopaminergic therapy, but also raised the possibility of a diagnosis of multiple system atrophy (MSA).
At the age of 59 years, the patient began to experience sudden, severe headaches with neck pain, tachycardia, sweating and pallor. The patient's blood pressure was markedly elevated, regardless of posture, during these episodes—an unusual finding given that he had no prior history of hypertension. In view of the array of symptoms, the patient was referred for medical evaluation of a suspected pheochromocytoma, a neuroendocrine tumor of the adrenal gland. He underwent 24-hour blood pressure monitoring, which showed no dip in blood pressure between day and night recording (average daytime blood pressure 111/75 mmHg; average night-time blood pressure 106/72 mmHg).
Selected biochemical test results from the patient are shown in Table 1. Initial screening tests revealed levels of plasma free metanephrines that were considered, by an endocrinologist, to be inappropriately elevated, a finding that is normally indicative of pheochromocytoma. In this case, the test results were deemed unreliable because the plasma free metanephrine level could have been falsely elevated as a result of treatment with a combination of levodopa, a catechol-O-methyltransferase (COMT) inhibitor and a monoamine oxidase (MAO) inhibitor. MRI of the abdomen showed a 1 
1 cm nodule in the left adrenal gland (Figure 1). A 123I-metaiodobenzylguanidine (123I-MIBG) single-photon emission CT (SPECT) scan showed an area of increased uptake in the left adrenal gland region (Figure 2), which corresponded with the nodule on the MRI scan and persisted over two consecutive days of imaging, strongly indicating an underlying pheochromocytoma. The 123I-MIBG scan also showed reduced cardiac uptake of the ligand, which was consistent with the postganglionic sympathetic denervation that is seen in PD (data not shown). A PET scan with 18F-fluorodeoxyglucose (FDG) performed 2 months later showed no abnormal radiotracer uptake in the region of the left adrenal gland and left perirenal space (data not shown). A positive 123I-MIBG scan for an adrenal pheochromocytoma with a subsequently negative 18F-FDG-PET scan suggested the possibility of a low-grade tumor.
Figure 1 | Abdominal MRI scan from a patient with a suspected pheochromocytoma.
A left adrenal mass (arrows) is visible in a | axial and b | sagittal noncontrast T1-weighted sequences. Abbreviations: P, posterior; R, right.
Figure 2 | Abdominal 123I-metaiodobenzylguanidine single-photon emission CT scan from a patient with a suspected pheochromocytoma.
a | Transverse, b | coronal and c | sagittal images show increased uptake of 123I-metaiodobenzylguanidine in the left adrenal gland (arrow) 24 hours postinjection.
The patient underwent a left adrenalectomy, which revealed adrenal medullary hyperplasia. The pheochromocytoma-like symptoms resolved completely following surgery. The final diagnosis was pheochromocytoma syndrome (that is, symptoms of pheochromocytoma secondary to adrenal pathology) due to adrenal medullary hyperplasia.
Discussion of diagnosis
Biochemical tests
Pheochromocytomas are rare (prevalence 0.1–0.6% in patients with hypertension), catecholamine-producing neuroendocrine tumors that originate primarily in the adrenal gland. The clinical presentation of pheochromocytoma is highly variable, making this condition a diagnostic challenge. The signs and symptoms include paroxysmal hypertension, headaches, episodic pallor or flushing, tachycardia, and feelings of panic or anxiety.1 The reported resolution of typical pheochromocytoma symptoms after surgical adrenalectomy in two patients with unilateral adrenal medullary hyperplasia suggests that pheochromocytoma could represent an initial pathological change that ultimately leads to formation of a tumor.2
Under normal circumstances, biochemical tests for excess of catecholamines and their metabolites form the mainstay of the initial diagnostic work-up for pheochromocytoma. Traditionally, these tests include measurements of plasma and urine catecholamines (epinephrine and norepinephrine), urine fractionated metanephrine and normetanephrine, urine vinylmandellic acid, and plasma free metanephrines and normetaneprine. If the tests are suggestive of a tumor, structural (CT or MRI) and functional (123I-MIBG SPECT or 18F-dopa PET) imaging techniques are then employed.1 Many variables, including drugs, physiological stimuli, and technical differences in measurement methodology, can spuriously elevate the levels of catecholamines and their metabolites, thereby producing false-positive results for pheochromocytoma.1
PD is a chronic neurodegenerative disorder that is characterized by dopaminergic dysfunction and disturbances in neurotransmitter systems, and is treated primarily with levodopa (3,4-dihydroxyphenylalanine). Levodopa is converted in the periphery to other catecholamines, such as norepinephrine, epinephrine and their metabolites, and treatment with this drug could, therefore, confound the interpretation of the biochemical tests that are used to diagnose pheochromocytoma.3 Indeed, false-positive diagnoses of pheochromocytoma have been shown to occur in patients with PD who are receiving levodopa.4, 5, 6 Consequently, the diagnosis of pheochromocytoma in the setting of concomitant PD pharmacotherapy requires investigative diligence, and, as discussed below, relies heavily on structural and functional imaging techniques.
In the absence of complicating factors, measurements of plasma free metanephrines or urinary fractionated metanephrines are the most sensitive tests for the diagnosis or reliable exclusion of pheochromocytoma.1, 7 Lenders et al. showed that the combination of different biochemical tests did not improve the diagnostic accuracy beyond that of a single test of plasma free metanephrines.7 Davidson et al. reported significantly higher levels of urinary dopamine, homovanillic acid, free normetanephrine, and free metanephrine in patients with PD who were receiving levodopa therapy, compared with a control group without PD and PD patients who were not receiving levodopa therapy. Moreover, the levels of these four analytes correlated with the daily dose of levodopa.3 Interestingly, in the patients who were receiving levodopa, the urinary catecholamine measurements were not significantly different from those obtained in controls (not on levodopa), suggesting that urinary catecholamine levels might not be confounded by levodopa administration,3, 4 and that measurement of urinary catecholamines in patients with PD who are receiving levodopa could be of some utility. The sensitivity of the test for urinary catecholamines is, however, generally lower than that of the test for plasma free metanephrines.
Dopamine agonist therapy might reduce the sensitivity of some of the biochemical tests used for the diagnosis of pheochromocytoma. For example, there are reports that bromocriptine and apomorphine decrease plasma levels of the catecholamine metabolite homovanillic acid.8 Box 1 provides a summary of the PD medications that could confound the results of biochemical tests for pheochromocytoma.
Patients with PD might have to be taken off their dopaminergic medications (for example, levodopa, dopamine agonists, MAO and COMT inhibitors) or switched to medications such as amantadine—which is thought not to confound the biochemical test results—before reliable biochemical assessment of catecholamine metabolites can be made. This approach is not always practical, however, especially in patients with moderate-to-advanced PD who rely heavily on levodopa for symptom control, and could experience worsening of symptoms such as hallucinations and confusion when switched to amantadine. In patients in the early stages of PD, on the other hand, it might be possible to withhold the medications for 
48 hours in order to yield cleaner biochemical test results.
Another biochemical test that might be useful in the diagnosis of pheochromocytoma is the measurement of plasma chromogranin A (CgA) levels. Grossrubatscher et al. showed that the CgA level had a sensitivity of 91% in identifying patients with pheochromocytoma, which increased to 100% when the test was combined with measurement of catecholamine levels. In addition, CgA levels have been shown to decrease after surgery for pheochromocytoma, and might be used to indicate whether the condition has been cured.9 At present, little is known about the efficacy of the CgA test in the context of PD and dopaminergic therapy.
A false diagnosis of pheochromocytoma is not unique to patients on levodopa therapy; there are also reports of pseudopheochromocytomas (symptoms suggestive of pheochromocytoma in the absence of adrenal pathology) in PD patients who are receiving selegiline.10, 11 Lefebvre et al. reported on one such patient who exhibited paroxysmal hypertension and other typical features of pheochromocytoma, and who showed high plasma norepinephrine and moderately elevated urinary vanillylmandelic acid concentrations, but had negative adrenal CT and 131I-MIBG scans. The authors believe that the symptoms occurred secondarily to drug interactions among selegiline, ephedrine and a tricyclic antidepressant.11 Hypertensive crises have also been reported in patients who are on a combination of selegiline and levodopa.12
Structural and functional imaging
Given the limited utility of biochemical tests in patients on dopaminergic medications, diagnostic imaging forms the mainstay of the initial work-up once pheochromocytoma is suspected in this context. CT and MRI are used to image the entire abdomen, including the pelvis, for the initial localization of the adrenal tumor.13 CT and MRI have comparable sensitivities and specificities, but T2-weighted MRI with gadolinium contrast is superior for detecting extra-adrenal tumors.1 123I-MIBG SPECT can improve on the poor specificity of CT and MRI for the diagnosis of pheochromocytoma. If 123I-MIBG is not available, 131I-MIBG can be used as an alternative, although it produces inferior imaging results to 123I-MIBG.1
123I-MIBG is a radioiodinated analog of norepinephrine that is transported actively into norepinephrine granules in sympathetic nerve terminals, and can consequently be used to study sympathetic denervation. 123I-MIBG SPECT is not routinely recommended in all patients with a biochemically indicated pheochromocytoma, but it is relevant in the present patient for two reasons. First, the biochemical tests could not be relied upon in this case. Second, 123I-MIBG scanning can also assist in confirming a diagnosis of PD and in differentiating idiopathic PD from other atypical parkinsonian syndromes such as MSA (a major component of which is autonomic dysfunction).14 Studies in patients with PD have shown reduced uptake of 123I-MIBG in myocardial sympathetic neurons, indicating impaired postganglionic sympathetic innervation. Other organs show normal 123I-MIBG uptake in patients with PD.15
The coexistence of PD and pheochromocytoma is a rare event, but one should, nevertheless, exercise caution when interpreting 123I-MIBG scan results in relation to diagnosing PD. Case reports have shown decreased myocardial 123I-MIBG uptake in untreated pheochromocytoma patients without PD. Uptake of the ligand improved markedly on surgical removal of the tumor. The decreased 123I-MIBG uptake in such cases could result from competition between increased levels of circulating catecholamines and 123I-MIBG.16 Medications such as tricyclic antidepressants, specific calcium antagonists and labetalol can also interfere with uptake of 123I-MIBG.17
If 123I-MIBG scanning is unavailable or produces negative results, PET scanning with 18F-fluorodopa, 18F-fluorodopamine or 18F-FDG can be used as an alternative.1 18F-FDG-PET is the most commonly available of these techniques, but it has low specificity for the diagnosis of a pheochromocytoma. 18F-fluorodopa and 18F-fluorodopamine offer comparable sensitivities to 123I-MIBG, especially in cases of malignant pheochromocytomas.1 In the present patient, the 123I-MIBG scan was positive but the 18F-FDG scan was negative, indicating that the tumor was probably low-grade in nature.
Discussion of treatment and management
Once a pheochromocytoma has been localized with the imaging techniques described above, elective surgical excision forms the mainstay of treatment. To keep catecholamine-induced surgical complications—such as hypertensive crises, cardiac arrhythmias, and ischemia and pulmonary edema—to a minimum, appropriate preoperative treatment is necessary.1 Pretreatment with an 
-adrenergic blocker (for example, phenoxybenzamine, prazosin or doxazosin) is started on an outpatient basis for 10–14 days before surgery. Other alternatives for preoperative management include labetolol and calcium channel blockers. Phenoxybenzamine is often the preferred choice because it blocks -adrenergic receptors noncompetitively, thereby preventing drug displacement from the receptors in case of excess catcholamine release during surgery. To ensure that the preoperative treatment has been adequate, several measurements, including blood pressure and cardiac monitoring, have been proposed.1 Currently, laparoscopic removal of the pheochromocytoma is the preferred surgical procedure.18 With adequate medical preparation, the operative mortality rate is less than 1% in the hands of an experienced anesthesiologist and surgeon.19 The two most common postsurgical complications are hypotension and hypoglycemia.1
All patients, regardless of PD status, should be followed by their physician for at least 10 years after surgery, and patients with extra-adrenal or familial pheochromocytoma should be followed up indefinitely.1 From the PD standpoint, PD medications should be continued right up to the time of surgery, and should be restarted as soon as possible after surgery in line with preoperative and postoperative oral intake guidelines.
Conclusions
The case described in this article illustrates the typical features of a pheochromocytoma and the diagnostic challenges inherent in diagnosing these tumors in the setting of PD. We have outlined the work-up of a patient in whom pheochromocytoma was suspected in the setting of comorbid PD, and in whom standard biochemical tests were confounded as a result of dopaminergic medications (Box 2). In addition to assisting in the localization of a pheochromocytoma, 123I-MIBG SPECT might also enable PD to be differentiated from conditions such as MSA in the setting of fluctuating blood pressure.
