F. Mantero
A.M. Masini G. Opocher
M. Giovagnetti
G. Arnaldi on behalf of the National Italian Study Group on Adrenal Tumors
Division of Endocrinology, Institute of Internal Medicine, University of Ancona, Italy
Key Words
Adrenal incidentaloma Adrenocortical cancer Epidemiology Pheochromocytoma
Abstract
Adrenal masses are more and more frequent- ly detected by adrenal ultrasound, computed tomography or nuclear magnetic resonance carried out for a reason other than the suspi- cion of adrenal disease (incidentalomas). The finding of an incidentaloma still leaves many diagnostic and therapeutic questions open. We report the results of a multicentric retrospective evaluation of patients with ad- renal incidentalomas, performed by a Study Group of the Italian Society of Endocrinolo- gy. According to the definition of incidenta- loma, exclusion criteria a priori were: severe or paroxysmal hypertension, frank hypo- kalemia and clinical signs of hypercortisol- ism or hyperandrogenism. 29 centers partici- pated in the study and the data obtained by questionnaire were collected in 2 centers for final elaboration. Center 1 carried out the epidemiological and clinical evaluation. Bas- al and dynamic hormonal evaluation of 786 among the 1013 cases recruited were per- formed in our center (center 2). Function- al studies included: diurnal rhythm of cor- tisol, urinary free cortisol (UFC), ACTH, DHEAS, 17-OH progesterone, testosterone, androstenedione, supine and upright plasma renin activity (PRA) and aldosterone, uri-
Supported in part by grant AIRC.
Adrenal Incidentaloma: An Overview of Hormonal Data from the National Italian Study Group
nary aldosterone, urinary catecholamines and VMA. The hormonal dynamic evalua- tion included the overnight dexamethasone suppression test (1 mg), CRH test and ACTH test. In our study, 89% (702 patients) of adrenal incidentalomas were non-hyper- secretory masses; 6.2% (49 patients) showed a preclinical Cushing’s syndrome (PCS) (at least two altered parameters of pituitary- adrenal axis); 3.4% (27 patients) were pheochromocytomas; 0.89% (7 patients) were aldosteronomas. One tumor was a mas- culinizing adrenocortical carcinoma. Two hundred sixty patients underwent surgical exploration and the histological diagnosis showed: 138 adenomas (53%), 32 carcino- mas (12%), 26 pheochromocytomas (10%), 16 myelolipomas (8%), 13 cystic lesions (5.5%), 7 tumors of neuronal lineage (3%), 12 metastases (4%), 13 others (5%). The 138 patients with adenomas had the following hormonal diagnosis: 103 nonfunctional ade- nomas (74%), 31 PCS (23%) and 4 cases of hyperaldosteronism (3%). In the patients with PCS an abnormal dexamethasone sup- pression test was found in 86% of cases (37/ 41 patients). Values for ACTH were low in 78% (32/41 patients). UFC was elevated in 64% of patients, the diurnal rhythm of corti- sol evaluated in 14 patients was absent in 7. Only in 50% of cases DHEAS values (12/24 patients) were decreased, whereas they were normal in the other 50%. Interestingly, 8 patients with normal DHEAS and normal UFC showed nonsuppressible cortisol by dexamethasone test (1 mg). Blunted ACTH response to CRH was detected in 9 of 14 patients (64%). Thus our data suggest that the best parameter for evaluating subclinical hypercortisolism seems to be the overnight
dexamethasone suppression test. In 27 pa- tients with pheochromocytoma 24-hour uri- nary catecholamine and VMA levels were elevated in 86 and 46% of cases respective- ly. In 7 patients with hyperaldosteronism upright PRA was suppressed in 100% of cases and aldosterone plasma levels were ele- vated in 6 patients (86%); serum potassium level was slightly decreased in 60% of cases. In 86 of 138 histologically proven adenomas, DHEAS levels were: normal in 59% of pa- tients, decreased in 36% and elevated in 4.6%, whereas in 22 of 32 cortical carcino- mas evaluated, DHEAS levels were normal in 63% of cases, decreased in 18% and ele- vated 18%. Post-ACTH 17-OH progesterone levels were elevated in 52% (62/118 pa- tients) of non-functioning adenomas and in 2 of 4 carcinomas. Not enough data are yet available postoperatively. In summary, en- docrine evaluation can lead to the identifica- tion of a nonnegligible number of cases of clinically unsuspected pheochromocytomas and subtle hypercortisolism (about 3.4 and 6.2%, respectively of all adrenal incidentalo- mas), while cases of primary subclinical al- dosteronism are rarely found. Most of the clinically silent carcinomas showed no major hormonal abnormalities, except for an in- creased DHEAS in 18% of cases. These data suggest that a series of selected biochemical tests should be performed in all incidentally discovered adrenal masses. This would in- clude 24-hour urinary catecholamine, plas- ma cortisol postsuppression with dexameth- asone 1 mg, upright PRA and DHEAS. The role of stimulated 17-OH progesterone in identifying the few patients with adrenal masses due to 21-hydroxylase deficiency re- mains to be evaluated.
KARGER
E-Mail karger@karger.ch Fax + 41 61 306 12 34 http://www.karger.ch
This article is also accessible online at: http://BioMedNet.com/karger
Franco Mantero. MD Clinica di Endocrinologia Ospedale Regionale Torrette I-60100 Ancona (Italy) Tel. +71 887061, Fax +71 887300, E-mail clendo@popesi.unian.it
Introduction
In the last few years, adrenal masses incidentally dis- covered, as a result of wider availability of abdominal imaging procedures (ultrasound, computed tomography (CT) or magnetic resonance imaging (MRI) have become a common and intriguing clinical problem. According to different series, the prevalence of unsuspected adrenal masses varies from 1 to 5% of patients undergoing CT [1-5]. Most of these masses are benign, but one should be aware that some may represent primary or metastatic malignancies [6-8].
The other major point to address is whether these tumors which in most cases are clinically silent are indeed secretory resulting in subclinical forms of Cushing’s syn- drome, pheochromocytoma, hyperaldosteronism or hy- perandrogenism [9-11]. With this in mind, correct diag- nostic and therapeutic strategies are of great clinical rele- vance. Not to be underestimated is the economic aspect of the problem, since an extensive endocrine workup is required to detect minor hormonal abnormalities; this probably causes more cost than benefit, especially if fur- ther imaging techniques are also required (e.g. MRI or radioisotopic scintiscans) [12, 13].
However, due to the great heterogeneity and the rela- tively low number of cases in each series as well as the lack of long-term follow-up, clear guidelines on the investiga- tion and management of adrenal incidentalomas have yet to be drawn up [1, 6].
In order to address these issues, we have set up a National Study Group on Adrenal Tumors under the patronage of the Italian Society of Endocrinology, with the purpose of organizing a multicentric collection of cases of adrenal incidentalomas. The study includes cases observed in 29 Italian centers, during the last 15 years, but is partially prospective, since most of the cases of the last 5 years have been studied according to a pre-determined protocol in operation in all to the participating centers.
National Italian Study Group on Adrenal Tumors
B. Ambrosi, E. Passini (Milano); A. Angeli, G. Osella, M. Terzolo, A. Ali (Torino); P. Bernini (Pisa); F. Bertolissi, F. Grimaldi (Udine); G. Borretta, GP Magro (Cuneo); M. Boscaro (Padova); F. Camanni, E. Mazza, De Fonzo (Torino); L. Capussotti, M. Dei Poli (Torino); F. Cavagnini, C. Invitti (Milano); E. Degli Uberti (Ferrara); A. Del Gaudio, G Solidoro (Bologna); M. Geravoglia (Novara); A. Giustina, P. Perini (Brescia); C. Letizia, A. Mancini (Roma); A. Liuzzi, M. Ghiggi, M. Torlontano (Foggia); V. Lo Cascio, P. Orsolini (Verona); P. Loli, F. Vignati (Milano); M. Mannelli, F. Tonelli (Firenze); M. Marugo, D. Bernasconi (Genova); E. Masenti, S. Rocca Rossetti (Torino); M. Messina, D. Flecchia, F. Morino (Torino); M. Muggeo, L. Furlani, R. Cas- tello (Verona); I. Portioli, R. Valcavi (Reggio Emilia); R. Rossi (Napoli); F. Santeusanio, S. Laureti (Perugia); R. Scardapane (Bari); A. Spinoglio (Ales- sandria); F. Virgili, M Riva (Mestre).
Patients and Methods
A multicenter, combined retrospective-prospective study involv- ing 29 different centers, most of which were endocrine-metabolic units of major university or regional hospital from all over Italy, was organized by a Coordinating Committee of the Study Group on Adrenal Tumors of the Italian Society of Endocrinology. The study, which required several meetings of the participants over the last 5 years, started in 1991 with the distribution of a consensus protocol for the investigation of subsequent cases of incidentaloma, and was followed by the distribution of an extensive questionnaire for the retrieval of data on all cases observed from 1980 to 1995. Data were collected in two centers (Torino and Ancona), for the final elabora- tion from the epidemiologic point of view (Torino) and endocrine aspects (Ancona). The software ‘Stat View 4.0’ for Macintosh was used for this purpose. Epidemiological and clinical data were ana- lyzed in center 1 (Torino), while basal and dynamic hormonal data were evaluated in our center.
By definition, an incidentaloma is an adrenal mass detected inci- dentally during an imaging workup performed for extra-adrenal com- plaints. According to this statement exclusion criteria a priori were: severe or paroxysmal hypertension, hypokalemia (≤3.5 mEq/l) and clinical signs of hypercortisolism or hyperandrogenism. Patients with malignancies known to frequently metastatize to the adrenals (lung and breast cancer, melanoma) were also excluded. At present, 1,013 cases have been recruited, but 227 were excluded from the hormonal evaluation because of inadequate data. Baseline hormonal data included diurnal rhythm of plasma cortisol, urinary free cortisol, (UFC), plasma ACTH, DHEAS, 17-OH Progesterone (17-OHP), tes- tosterone, supine and upright plasma renin activity (PRA) and aldo- sterone, urinary aldosterone and catecholamine excretion and vanil- lylmandelic acid (VMA). The hormonal dynamic tests included an overnight 1 mg dexamethasone suppresssion test, a CRH test and a short ACTH stimulation test. The hormonal data were recorded in the questionnaire both as raw data and classified in one of the follow- ing categories: low, normal, high; reference values of each center were also provided. The same methodology was applied for the classifica- tion of the response to the functional tests. Both baseline and dynamic hormonal data were only available in a variable proportion of cases.
Results
Among the 786 incidentalomas analyzed, 702 were classified, on the basis of clinical and biochemical data, as nonhypersecreting adrenal masses (89%), 49 as preclini- cal Cushing’s syndrome (6.2%), (when 2 or more tests of hypothalamus-pituitary-adrenal (HPA) axis function were abnormal); 27 were pheochromocytomas (3.4%), 7 were aldosterone-producing adenomas (0.89%) and 1 a virilizing tumor (fig. 1a).
Pathology was available in 260 patients who under- went surgery, with the following histological picture: 138 cortical adenomas (53%), 32 cortical carcinoma (12%), 26 pheochromocytomas (10%), 16 myelolipomas (8%), 13 cysts (5.5%), 12 metastases (4%), 7 ganglioneuromas (3%)
HYPERALDOSTERONISM 0.89%
Metastases
Cysts 5%
4%
Neuronal tumor 3%
Myelolipoma 8%
5% Other
PHEOCHROMOCYTOMA 3.4%
PHEO 10%
PRE CUSHING’s Syndrome 6.2%
Carcinoma 12%
Adenoma 53%
80
74%
70
60
NONHYPERSECRETORY 89%
%Patients
50
40
23%
30
20
3%
10
a
0
Nonhypersec
b
Pre-Cushing’s syn
Conn
and 13 (5%) other pathology (fig. 1b). In particular, among the 138 adrenocortical benign adenomas who un- derwent surgery 74% were non-hypersecretory, 23% were secreting cortisol in slight excess, and 3% secreted aldoste- rone. The total number of benign cortical adenomas, on the basis of only clinical and morphological aspects, was 680. The endocrine evaluation of these patients has shown alterations in the HPA axis as follows: 33.9% (94/ 277) had low ACTH levels; 19% (27/142) had an abnor- mal circadian rhythm of plasma cortisol; 17.7% (58/326) had UFC levels above normal; in 16% (56/349) cortisol was not adequately suppressed by dexamethasone, and in 24% (25/103) ACTH was hyporesponsive to CRH stimu- lation. As previously reported, 49 patients met the criteria of 2 or more abnormal parameters to be classified as pre- Cushing’s syndrome. The prevalence of each of these abnormalities in patients with the final diagnosis of pre- Cushing’s syndrome is reported in figure 2. It is worth
noting that, while most of the patients with UFC above normal had also a lack of suppression after dexametha- sone, the latter occurred also in several patients with nor- mal UFC.
Hormonal data in patients with adrenocortical carci- noma were scanty; only UFC was available in about 40% of the cases and above normal in one-third of them.
Low levels of DHEAS were found more frequently in patients with tumors of adrenocortical origin than in those with other etiology (metastasis, cysts, ganglioneuro- ma, myelolipoma): 54% vs. 24% respectively. In 378 patients with benign cortical adenoma, DHEAS levels were normal in 56% (212), low in 36% (137) and high in 7.6% (29), while in 22 carcinomas it was normal in 63% (14), low in 18% (4) and high in 18% (4) (fig. 3). Baseline values of 17-OHP were elevated in 7% (22/300) of the patients with adenomas and in 2 of the 7 patients tested with carcinomas. After ACTH stimulation test, an in-
100
37/41
32/42
80
9/14
25/41
7/14
12/24
% of patients
60
40
20
0
Dex1mg ACTH CRH test UFC Rhythm DHEAS no suppr.
low
hypores. ab. normal absent
low
2
100
14/22
% of patients
80
212/378
60
137/378
40
4/22
4/22
20
29/378
0
normal
low
high
Cortical adenoma n = 378
Cortical carcinoma n = 22
3
100
13/15
% of patients
80
15/27
60
6/15
40
20
0
Urinary Catecholamines
VMA
Hypertension mild / moderate
creased (≥5 ng/ml) response of 17-OHP was found in 52% (62/118) of the patients with adenomas and in 3 of 4 patients with carcinomas; cortisol was normal or even hyperresponsive (36%) at variance with what one would expect in classic forms of congenital adrenal hyperplasia (21-OH deficiency). Furthermore, in only 1 case out of 27 there was an excessive response of ACTH to CRH stimulation, while in 6 the response was blunted. Finally, the percent of the cases with hyperresponsive 17-OHP was similar in patients with bilateral or unilateral masses. Even half of the patients with pre-Cushing’s syn- drome showed an increased response of 17-OHP to ACTH. (Only few data in this regard are so far available postoperatively; preliminary results demonstrate the dis- appearance of the exaggerated 17-OHP response in most cases).
In 27 patients a pheochromocytoma was found at sur- gery; whilst only 15 of them had mild to moderate hyper-
tension, none had symptoms of enhanced adrenergic dis- charge. Total urinary catecholamines were elevated in 13/ 15 patients (sensitivity 86%, specificity 92%), and VMA in 6/15 patients (sensitivity 46%, specificity 92; fig. 4).
All 7 patients with primary aldosteronism were moder- ately hypertensive, but had serum K concentration higher than 3.5 mEq and were not suspected of having a minera- locorticoid form of hypertension. Upright PRA levels were suppressed in 100% of cases, plasma aldosterone was apparently within normal limits in 1 case and slightly increased in the other 6. Sensitivity for PRA resulted 100%, while specificity was 63%; sensitivity for plasma aldosterone was 75%, and specificity was 81%.
Discussion
The analysis of the hormonal data collected in this large multicentric study confirms that an endocrine workup of patients with incidentally discovered adrenal masses might lead to the recognition of a substantial num- ber of preclinical or subclinical hormone-producing tu- mors resulting in an advantageous early cure. Recent pro- gress in surgical techniques (especially the laparoscopic approach) has resulted in a great interest in the early diag- nosis of these diseases with the lower risk of this treatment [14, 15].
The most frequently diagnosed functioning tumors were those producing cortisol (6.2%) and catecholamines (3.4%), while hyperaldosteronism was rarely diagnosed (less than 1%) and androgen-secreting neoplasm hardly ever (1 case). The endocrine investigation required to make a diagnosis of ‘preclinical Cushing’s syndrome or pre-Cushing’s syndrome’ is still a matter of debate [9, 16- 20]. We have chosen the criteria of obtaining at least two or more abnormal results from the classical tests of the HPA axis. The most frequently found abnormal parame- ters were nonsuppressible cortisol and low baseline ACTH levels; however, due to the lack of specificity of the latter (several commercially available assays do not dis- criminate low from normal levels) we have required fur- ther tests in most cases to make a presumptive diagnosis of pre-Cushing’s syndrome. However, the existence of other clinical or biochemical markers of slight hypercorti- solemia were not apparent, since the high prevalence of obesity, diabetes and hypertension found in patients with incidentaloma was not significantly greater in those with pre-Cushing’s syndrome. The same holds true for other parameters of glucocorticoid activity such as osteocalcin, which has been found to be lower in patients with inciden- taloma irrespective of its apparent functional activity [10]; similarly, even the unilateral concordant uptake of radiocholesterol has not been demonstrated to correlate with the hormonal functional activity of the mass [6].
Our data suggest that subtle disturbances of steroid secretion are probably present independently of our clini- cal ability to detect them. However, even if these data are not provided by the present study, we have little informa- tion on the possibility of an evolution towards overt Cush- ing’s syndrome in some cases of incidentaloma. The pic- ture is even more complex if one takes into consideration the other relatively more frequent hormonal abnormality found in these patients, i.e. the increased levels of 17- OHP and its exaggerated response to ACTH, which was found in about half the patients with cortical tumors in
our study. This again was not related to the pattern of cor- tisol secretion, or to the level of plasma androgens. Fur- thermore, no increased ACTH response to CRH was found as one would expect in case of even subtle 21- hydroxylase deficiency. Altogether, even if postoperative data are not included in the present study, it appears quite likely, in agreement with other recent observations, that this peculiar steroid pattern is not a marker of a congenital enzymatic disorder evolving in adrenal nodules (as it may indeed occur in true 21-hydroxylase deficiency) but rather a sign of a disturbed intratumoral steroidogenesis [9, 21, 22]. A molecular analysis of the 21-hydroxylase gene is required to identify the few cases with primary genetic defects.
Similar considerations may also apply to the frequent finding (~ 40%) of low levels of DHEAS in adrenocortical adenomas; this again could indicate a disturbed intra- adrenal enzyme activity (for example, impaired 17-20- lyase) rather than to be ascribed, as previously suggested, to a suppressed ACTH due to mild hypercortisolism [9, 11]. In fact, low DHEAS levels were not inversely related to cortisol levels in our study. DHEAS levels were not found to be useful in distinguishing between benign and malignant adrenal tumors, even if the percentage of ele- vated values was higher in the latter group [23].
Clinically silent pheochromocytoma was the second most prevalent form of hyperfunctioning tumor, ap- pearing in 3.4% of all cases and in 10% of the cases who underwent surgery. In fact, half of these patients had some degree of hypertension but without the clinical character- istics which would have led us to suspect the presence of a catecholamine-secreting tumor. Urinary total catechol- amines were increased in almost all cases in which they were measured and this finding has an important impact on the diagnostic approach to the adrenal mass. An early diagnosis of pheochromocytoma is important to avoid the likelihood of future hypertensive crisis, and also because a preoperative specific diagnosis is essential in decreasing the risk of complications at operation as well as in diving needle biopsy [24-26]. The diagnostic role of MIBG or Octeotride scintiscan was not addressed in the present study.
The most infrequent form of hormonally active adre- nal tumors within this series was primary aldosteronism. This may be due to the exclusion criteria of our study (hy- pokalemia, moderate to severe hypertension); however, it is now well known that approximately one-third of these patients may be normokalemic, though hypertension is a consistent finding (at variance with pheochromocytoma).
In conclusion, endocrine evaluation of adrenal inci- dentaloma can lead to the identification of a significant number of cases of subtle hypercortisolism and clinically unsuspected pheochromocytoma, while cases of primary subclinical aldosteronism are rarely found. Most cases of clinically silent carcinoma showed no major hormonal abnormalities except for an increase in DHEAS in less than 25% of cases. These data suggest that a series of
selected biochemical tests should be performed in all inci- dentally discovered adrenal masses. This should include at least a postdexamethasone plasma cortisol, 24-hour urinary catecholamine, upright PRA and DHEAS, while the role of post-ACTH 17-OHP to identify rare subclini- cal forms of 21-OHase deficiency remains to be eluci- dated.
References
1 Gross MD, Shapiro B: Clinical review 50. Clin- ically silent adrenal masses. J Clin Endocrinol Metab 1993;77:885-888.
2 Mitnick JS, Bosniak MA, Megibow AJ, Nadich DP: Nonfunctioning adrenal adenomas discov- ered incidentally on computed tomography. Radiology 1983;148:495-499.
3 Moulton JS: CT of the adrenal glands. Semin Roentgenol 1988;33:288-303.
4 Abecassis M, McLoughin MJ, Langer B, Kud- low JE: Serendipitous adrenal masses: Preva- lence, significance and management. Am J Surg 1985;149:783-788.
5 Herrera MF, Grant CS, Van Heerden JA, Shee- dy PF, Ilstrup DM: Incidentally discovered ad- renal tumors: an institutional perspective. Sur- gery 1991;110:1014-1021.
6 Kloos RT, Gross MD, Francis IR, Korobkin M, Shapiro B: Incidentally discovered adrenal masses. Endocr Rev 1995;16:460-484.
7 Lewinsky BS. Grigor KM, Symington T, Ne- ville AM: The clinical and pathologic features of ‘non-hormonal’ adrenocortical tumors. Can- cer 1974;33:778-790.
8 Falke TH: Imaging of adrenal metastases. Curr Opin Radiol 1991;3:681-686.
9 Griffing G: A-I-D-S: The new endocrine epi- demic (editorial comment). J Clin Endocrinol Metab 1994;79:1530-1531.
10 Ambrosi B, Peverelli S, Passini E, Re T, et al: Abnormalities of endocrine function in pa- tients with clinically ‘silent’ adrenal masses. Eur J Endocrinol 1995;132:422-428.
11 Terzolo M, Osella G, Ali A, Borretta G, Magro G, Termine A, Paccotti P, Angeli A: Different patterns of steroid secretion in patient with adrenal incidentaloma. J Clin Endocrinol Me- tab 1996;81:740-744.
12 Ross NS, Aron DC: Hormonal evaluation of the patient with an incidentally discovered ad- renal mass. N Engl J Med 1990;323:1401- 1405.
13 Copeland PM: The incidentally discovered ad- renal mass. Ann Intern Med 1983;98:940- 945.
14 Guazzoni G, Montorsi F, Da Pozzo L, Rigatti P. Lanzi R, Pontiroli AE: Transperitoneal lapa- roscopic versus open adrenalectomy for benign hyperfunctioning adrenal tumors: a compara- tive study. J Urol 1995;153:1597-1600.
15 Go H, Takeda M, Imai T, Komeyama T, Ni- shiyama T, Morishita H: Laparoscopic adre- nalectomy for Cushing’s syndrome: Compara- tion with primary aldosteronism. Surgery 1995;117:11-17.
16 Charbonnel B, Chatal JF, Ozanne P: Does the corticoadrenal adenoma with pre-Cushing syn- drome exist ? J Nucl Med 1981;22:1059-1061.
17 Oelkers W: Diagnostic puzzle of the adrenal incidentaloma (invited commentary). Eur J Endocrinol 1995;132:419-421.
18 Reincke M, Nieke J, Krestin GP, Saeger W, Allolio B, Winkelman W: Preclinical Cushing’s syndrome in adrenal ‘incidentalomas’: Com- parison with adrenal Cushing’s syndrome. J Clin Endocrinol Metab 1992;75:826-832.
19 Ross NS: Epidemiology of Cushing’s syndrome and subclinical disease. Endocrinol Clin North Am 1994;23:539-546.
20 Caplan RH, Strut PJ, Wickus GG: Subclinical hormone secretion by incidentally discovered adrenal masses. Arch Surg 1994;129:291-296.
21 Seppel T, Schalghecke R: Augmented 17 hy- droxylase activity in patients with incidentally discovered adrenal tumours (‘incidentalomas’). Clin Endocrinol (Oxf) 1995;41:445-451.
22 Del Monte P, Bernasconi D, Bertolazzi L, Meozzi M, Badaracco B, Torre R, Marugo M: Increased 17alfa-hydroxyprogesterone re- sponse to ACTH. Clin Endocrinol 1995;42: 273-277.
23 Osella G, Terzolo M, Borretta G, et al: Endo- crine evaluation of incidentally discovered ad- renal masses (incidentalomas). J Clin Endocri- nol Metab 1994;9:1532-1539.
24 McCorkell SJ, Niles NL: Fine needle aspiration of catecholamine-producing adrenal masses: A possible fatal mistake. Am J Roentgenol 1985; 145:113-114.
25 Casola G, Nicolet V, vanSonnenberg E, With- ers C, Bretagnolle M, Saba RM, Bret PM: Un- suspected pheochromocytoma: risk of blood- pressure alterations during percutaneous adre- nal biopsy. Radiology 1986;159:733-735.
26 Bravo EL: Evolving concepts in the pathophys- iology, diagnosis and treatment of pheochro- mocytoma. Endocr Rev 1994;15:356-368.