| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Special Articles |
Departments of Hypertension, Urology, and Medical Informatics, Hôpital Broussais-Saint Joseph, 75908 Paris Cedex 15; and Gestion Essais Cliniques, Etudes Statistiques Monitoring, 92120 Montrouge, France
Address all correspondence and requests for reprints to: Dr. Pierre-François Plouin, Hypertension Unit, Hôpital Européen Georges Pompidou, 20 rue Leblanc, 75908 Paris Cedex 15, France. E-mail: pierre-francois.plouin{at}egp.ap-hop-paris.fr
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
As preoperative management, anesthesia, and surgery may be complex and involve large and acute variations in BP and heart rate, patients with pheochromocytoma should be referred to centers with extensive experience of the disease (1, 2). Little is known, however, about the rate of perioperative complications of pheochromocytoma surgery. We searched the literature published over the last 10 yr for series with at least 50 pheochromocytoma operations. We retrieved 9 reports quantifying operative mortality and morbidity (3, 4, 5, 6, 7, 8, 11, 12, 13) (if several reports were published from the same institution, we included only the most recent report). The reported mortality rates were 0–6.7%, and the morbidity rates were 3.3–36.1%. There are several plausible explanations for this wide range, including referral biases, incomplete reporting, and ambiguities in the definition of complications.
We report preoperative and postoperative complications in a large series of consecutive patients with pheochromocytoma referred to a single center. Preoperative management was assumed by a single physician (P.F.P.), and a single surgeon (J.M.D.) operated on all patients. To limit reporting biases, we obtained an external review of the medical, anesthetic, and surgical records of all operations performed in this center on patients with pheochromocytoma. Complications were defined before data collection. To detect risk indicators for complications, we analyzed the clinical, biological, and tumoral features of the patients who suffered complications.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The procedures used for pheochromocytoma diagnosis were
consistent with institutional guidelines and have been described
previously (9, 14, 15, 16, 17). Family history and phenotypic
signs of multiple endocrine neoplasia type 2 (MEN 2), von Hippel-Lindau
disease, and neurofibromatosis 1 were analyzed to identify any
underlying familial disease (9, 18). Histological evidence
of pheochromocytoma was obtained in all patients. BP was determined
using a mercury sphygmomanometer after patients had spent a mean of 15
min in the horizontal position. Patients were hospitalized in the
hypertension department for imaging studies and preparation for
intervention, including adaptation of antihypertensive treatment if
necessary. They were transferred to the urology department on the day
before the operation. The mean of two consecutive readings obtained
immediately before transfer to the Department of Urology was taken as
the preoperative BP determination. Patients with sustained hypertension
were mostly given 
- and ß-adrenergic antagonists before surgery.
Antihypertensive medication was given to lower BP to 140/90 mm Hg or
lower. Anesthesia was achieved using a combination of flunitrazepam,
fentanyl, and thiopental, and curarization was achieved using
vecuronium. Radial artery pressure, pulmonary artery pressure, and
electrocardiograph were monitored continuously. Infusions of
nitroprusside, nicardipine, and antiarrhythmic agents were prepared in
advance and administered as soon as necessary. Controlled ventilation
was established, enabling a positive end expiratory pressure to be
achieved if pulmonary wedge pressure increased. Volume replacement was
determined by pulmonary wedge pressure levels. Our routine operation
technique involved a median transperitoneal approach, making it
possible to explore both adrenal glands, the organ of Zuckerkandl, the
urinary bladder, and the periaortic and pericaval lymphatic chains.
Left adrenalectomy was performed after medial visceral rotation of the
spleen and the tail of the pancreas. In this series, extending from
1975 to 1997, we used laparoscopic adrenalectomy in only a few cases
with MEN 2 and small adrenal tumors. Tumor extirpation involved the
adjacent adrenal gland (in cases of adrenal tumors) and/or any other
possible tumors, lymph nodes, or visceral metastases.
Clinical, tumoral, and complication criteria
Sustained hypertension was diagnosed in patients with a BP of 140/90 mm Hg or more in the absence of paroxysmal symptoms and in those taking antihypertensive drugs. Hyperglycemia was defined as two plasma glucose assay results above 7 mmoL/L (19). Pheochromocytoma site and size were confirmed at surgery. Malignancy was defined as histological evidence of tumor cells at sites at which chromaffin tissue is not normally present and/or evidence of distant metastases documented by computed tomographic scan or scintigraphy (9, 20, 21). Recurrence was defined as the reappearance of disease after complete tumor eradication confirmed by negative biochemical and imaging tests (9). A malignant recurrence was defined according to the malignancy criteria described above; other recurrences were deemed benign. In addition to reinterventions for tumor persistence or recurrence, some reinterventions were required for the resection of new tumors affecting the contralateral adrenal gland or ectopic chromaffin tissue in patients with familial pheochromocytoma.
Complications were defined before data collection and analysis. Any adverse event occurring between the induction of anesthesia and the 30th day after the operation that was life threatening, resulted in death or permanent or substantial disability, or was associated with extended hospitalization was considered to be a perioperative complication. Complications included mortality from any cause, nonfatal cardiovascular events, thromboembolism, resection of a neighboring organ, further surgery, and any infectious or incision problem extending hospitalization in the surgical department beyond ten days. We also documented episodes of hypoglycemia that did not fulfill the above criteria for complication.
Study population and data collection
Between September 1975 and December 1997, pheochromocytoma was diagnosed in 164 patients referred to the Department of Hypertension. This list of patients was merged with that of patients operated on for pheochromocytoma in the Department of Urology during the same period. Eleven patients later underwent surgery elsewhere at the request of their personal physician or surgeon, and another 6, who had previously undergone surgery elsewhere, were referred for nonsurgical management of a malignant pheochromocytoma. Therefore, the records of 147 patients were available for analysis of perioperative outcome. Sixteen of 23 patients with malignant pheochromocytoma and 14 of 124 with benign pheochromocytoma had persistent tumors, tumor recurrence, or new tumors requiring surgery. Sixteen operations had been performed before referral. Therefore, 165 operations were analyzed. An independent clinical research assistant reviewed the medical, anesthetic, and surgical records of all these patients and input relevant perioperative data into a computer file.
Data analysis
Differences between means were assessed using t test
or Mann-Whitney test as appropriate. The 
2
test or Fisher’s exact test was used to compare qualitative variables.
We looked for univariate associations between the presence or absence
of complications, the year of the operation, the number of previous
operations if any, and the following patient and tumor characteristics:
age, sex, history of cardiovascular events, the presence of an
underlying familial disease, sustained hypertension or hyperglycemia,
preoperative BP levels, urinary metanephrine excretion, plasma
catecholamine concentration, tumor status (benign or malignant), and
tumor site and size. We could not test the statistical relationship
between the risk of complications and the surgical technique because
the same surgical approach was used in almost all the patients.
Variables significantly associated with the presence of complications
were entered in a logistic regression model to determine which
variables were independently associated with the risk of complication.
Data were analyzed with StatView 5.0 statistical software (SAS Institute, Inc., Cary, NC). P < 0.05 was
considered significant.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The characteristics of the patients at first operation are
summarized in Table 1
. The patients were
13–82 yr old. Pheochromocytoma was associated with a familial disease
in 31 patients. Hypertension was sustained in 129 (87.8%) patients,
and 46 (31.3%) were hyperglycemic. Fifteen patients had a personal
history of cardiovascular events (myocardial infarction, 8; stroke, 7).
Pheochromocytoma at first operation was in the right adrenal gland in
79 patients and in the left in 42. It was bilateral in seven patients
and extraadrenal in 19: in the Zuckerkandl body (n = 10), renal
hilum (n = 5), prostate (n = 2), mediastinum (n = 1), or
pericardium (n = 1). Twenty-three patients, including 6 with
extraadrenal tumors, had malignant pheochromocytoma at the first
operation. Malignant tumors were significantly larger than benign
tumors (Table 1). Surgery was performed using a median transperitoneal
approach in 156 operations, a flank incision in 4 operations, and
thoracotomy in the 2 cases of thoracic pheochromocytoma, and 3 patients
with MEN 2 and small adrenal tumors underwent laparoscopic surgery.
|
Thirty patients needed one or more reinterventions for a
persistent, recurrent, or new tumor 91 ± 73 months (range, 1–231
months) after the initial operation (Table 1
). The proportion of
patients needing reintervention was higher in cases with a malignant
tumor at first operation than in those with benign tumor. Nine patients
with sporadic, initially benign tumors developed a true recurrence
requiring surgery. Eight of these recurrences were malignant.
Reintervention was also needed for the occurrence of a new tumor in 7
of the 31 patients with familial disease. All new tumors were
benign.
Complications
Patients with complications are listed in Table 2, according to benign/malignant status
and in ascending order of tumor diameter. Perioperative mortality was 4
of 165 or 2.4%. Patient 7, with a history of asthma and benign
pheochromocytoma, died with an abrupt collapsus without arrhythmia on
induction of anesthesia, presumably of anaphylactic shock. Another 3
patients with large malignant tumors also died. Patient 22 died of
pulmonary edema during the operation. Patient 24 had undergone an
initial adrenal resection and bilateral nephrectomy 1 yr earlier for
malignant pheochromocytoma and bilateral renal cell cancer; he died
suddenly during hemodialysis within 1 day of reintervention. Patient 26
had MEN 2 with bilateral malignant pheochromocytoma; a liver metastasis
biopsy led to postoperative bleeding, and the patient died during a
subsequent emergency operation.
|
and the 11
patients in Table 3) underwent
splenectomy, and patient 6 had a splenic hematoma (see Table 2).
Patients undergoing splenectomy had pheochromocytoma arising on either
the left or the right side, either benign or malignant, with a tumor
diameter of 30–140 mm (Table 3). No postoperative hernia was reported
in this series of patients, who, in most cases, underwent surgery via
midline incision.
|
Perioperative features for operations with complications were
compared with those for operations without complications (Table 4). There were no significant differences
in the numbers of patients with familial disease or a history of
previous cardiovascular disease, the age and sex of the patients, the
median year of operation, or the proportion of patients with
hyperglycemia or sustained hypertension. Preoperative treatment score,
defined as the number of antihypertensive agents administered, did not
differ between cases with and without complications; nor did the
proportion of patients given the -adrenergic antagonist prazosine, a
ß-blocker, or another antihypertensive agent (see Table 4).
Preoperative systolic BP was significantly higher in operations with
complications that in those without complications. Mean urinary
metanephrine excretion in operations with complications was twice that
in those without, but there was no difference in plasma catecholamine
concentration. Complications occurred more frequently in operations on
patients who had undergone a previous pheochromocytoma operation and in
those involving patients with a malignant tumor. Tumor size and site
were not related to the occurrence of complications.
|
),
but not to tumor status. The same logistic model was applied to spleen
complications (n = 13), and only preoperative systolic BP was
significantly associated with the occurrence of splenectomy or of
splenic hematoma [odds ratio (OR), 1.189; 95% confidence interval,
1.003–1.409; P = 0.046]. Finally, we analyzed
predictors of perioperative complications in first operations and in
repeat operations. The ORs for the risk of complications associated
with high urinary metanephrine excretion and with high preoperative
systolic BP were very similar for all operations, first operations, or
repeat operations (see Table 5).
|
Twenty-five operations were followed by episodes of hypoglycemia requiring the infusion of hypertonic glucose solution. There was no significant difference in the proportion of patients with preoperative hyperglycemia [8 of 25 (32.0%) vs. 43 of 131 (32.8%)], malignant pheochromocytoma [3 of 25 (12.0%) vs. 31 of 131 (23.7%)], or in preoperative plasma catecholamine concentrations between cases with and without hypoglycemia. Sixteen operations were followed by adrenal insufficiency. This was an expected consequence of adrenal resection as all affected patients had undergone resection of both adrenal glands for bilateral (synchronous or nonsynchronous) and/or malignant pheochromocytoma, except 1 patient who had a benign pheochromocytoma and ectopic ACTH syndrome. This previously reported patient (22) presented with Cushing disease and suffered transient postoperative adrenal insufficiency.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The strengths of our study are the external review of a large series of records from consecutive patients, with benign, recurrent, and malignant pheochromocytomas prepared for surgery in a single unit and operated on by the same surgeon, and the definition of complications before data collection and analysis. Previous reports of perioperative complications in pheochromocytoma surgery were self-audits, were limited to mortality or preoperative complications, and/or did not define the criteria for the presence of complications beforehand. They may therefore have underestimated the true prevalence of complications (23, 24, 25). The weaknesses of our study are its retrospective nature and the consequences of the a priori definition of complications. Mild complications, such as wound infections and urinary tract infections of less than 10-day duration, were rarely identified and probably underreported in patients’ records. In contrast, some operations involving resection of a neighboring organ (nephrectomy during resection of a left renal hilum tumor in patient 10 and cholecystectomy in patient 19) were considered complicated according to predefined criteria, whereas the surgeon considered these resections appropriate to ensure complete tumor excision or prevention of postoperative cholecystitis. Exclusion of these two cases would not have altered the overall results of the analysis, however. This series of operations performed from 1975 to 1997 mostly concerns the era of open surgery. A lower frequency of complications, a shorter hospital stay, and a lower level of postoperative scars are expected from laparoscopic surgery. Our results cannot be used, however, for historical comparison between open and laparoscopic pheochromocytoma surgery, because a selection bias would be introduced by the selection for open surgery of patients with large, recurring, or malignant pheochromocytomas, whereas patients selected for laparoscopic surgery usually have nonmalignant tumors of 6 cm or less (26, 27, 28).
One unexpected finding was the high incidence of spleen complications, 13 of 165 (7.9%), including 12 splenectomies and 1 splenic hematoma. Spontaneous spleen rupture and preoperative spleen complications have been reported in patients with pheochromocytoma (29, 30, 31, 32), but without quantification. In this series, spleen complications were observed in patients with pheochromocytomas on either the left or right side and were not related to tumor size. The operative strategy used, an open midline incision with routine exploration of both adrenal glands and of the area around the aorta and vena cava, probably contributed to the high frequency of splenectomies. In multivariate analysis, the risk of spleen complications was also associated with the level of preoperative systolic BP. Spleen capacitance is altered by acute changes in BP and plasma catecholamine concentrations (33, 34). Poor BP control may lead to acute changes in spleen volume and to a higher susceptibility to spleen trauma during abdominal exploration and spleen mobilization. In recent years improvements in preoperative tumor location have made possible less extensive abdominal exploration during pheochromocytoma surgery, and more attention has been paid to the long-term consequences of splenectomy (35). This may explain why spleen complications occurred in only three cases in the last 10 yr. The excellent preoperative localization studies that are possible nowadays have rendered general exploration unnecessary, and indeed such exploration is no longer justified.
All operations were performed by one of us (J.M.D.), using a median transperitoneal approach in 156 of 165 (95.4%) cases, and the median year of operation did not differ between operations with or without complications. Consequently, the incidence of complications could not be related to the surgeon’s experience, and it was not possible to test the statistical relationship between the risk of complication and the operation technique used. Only 3 patients underwent laparoscopic adrenalectomy. These patients had small adrenal tumors detected by screening in a family with NEM 2. This small number of patients precluded any comparison between the transperitoneal and the laparoscopic approaches. Complication rates were similar after resection of adrenal or extraadrenal pheochromocytomas. The overall incidence of complications was related to preoperative urinary metanephrine excretion and systolic BP and was much higher at reintervention than at first intervention.
Urinary metanephrine excretion and systolic BP were independent predictors of complications. In multivariate analysis, urinary metanephrine excretion was associated with complications after all (OR, 1.184; P = 0.002), first (OR, 1.172; P = 0.007), and repeat (OR, 1.157; P = 0.237) operations. Although P values were increasing because of decreasing statistical power, the OR for the association between preoperative systolic BP and complications was also very consistent, yielding 1.138/10 mm Hg for all operations (P = 0.052), 1.138 for first operations (P = 0.076), and 1.338 for repeat operations (P = 0.119). Metanephrine excretion is positively related to catecholamine turnover (36) and tumor size (9) and is much higher in patients with malignant pheochromocytoma than in those with a benign tumor (9). We found no relationship between plasma catecholamine concentrations and the incidence of complications, probably because plasma catecholamine levels are highly variable in patients with pheochromocytoma, whereas determination of metanephrine excretion provides an integration of pheochromocytoma metabolic activity over a 24-h period (14). Preoperative systolic BP was significantly related to the incidence of complications. This relationship does not necessarily imply that adequate BP control can prevent complications, because resistant hypertension may be an independent marker of cases at high risk. The relationship was independent, however, of plasma catecholamine levels, urinary metanephrine excretion, and tumor size and status (benign/malignant). High BP levels and, more specifically, the permanent component of hypertension may be reduced by antihypertensive agents in subjects with pheochromocytoma (11, 16, 37). We therefore suggest that BP should be normalized whenever possible before pheochromocytoma surgery to reduce the incidence of complications. Competitive adrenergic antagonists such as prazosine are frequently used as the first-line treatment because they do not expose the patient to the risk of postoperative hypotension (11, 16). Longer acting antagonists, such as phenoxybenzamine and/or the catecholamine synthesis inhibitor, metyrosine, may be used in patients who are difficult to manage (37).
The incidence of complications after repeat intervention was much higher than that after the initial pheochromocytoma resection (OR, 5.29). Fibrosis after previous operations; extended dissection of the adrenal, periaortic, or pericaval sites of new tumors or lymph node metastases; and/or the presence of metastases extending into neighboring organs all account for the high risk of complication at reintervention. In this series we considered that subjects at high risk should be screened for recurrence (9) and that focal recurrences (mostly lymph node recurrences) were indications for reintervention. There is no evidence, however, that the early detection and resection of recurrences prolong survival (38). Given the high frequency of complications after reintervention, the treatment indications and methods should be discussed with an experienced surgeon after considering conservative management. Nonsurgical pheochromocytoma management by chemotherapy, the application of therapeutic doses of meta-iodobenzyl guanidine, or both (39, 40) is at best palliative, however. Medication with metyrosine with or without phenoxybenzamine (37) or somatostatin analogs (41, 42) may help control BP and alleviate symptoms in patients with pheochromocytoma, but has no antiproliferative effect.
In conclusion, physicians and patients should be informed that surgery for pheochromocytoma consistently involves a risk of complications, including spleen damage. Careful BP control probably helps prevent complications. Patients with high secretion tumors and those undergoing repeat intervention, generally for malignant recurrence, are at high risk of complications and should be referred to centers familiar with surgical and nonsurgical pheochromocytoma management.
|
Footnotes |
|---|
Received June 15, 2000.
Revised October 24, 2000.
Revised December 22, 2000.
Accepted January 6, 2001.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  K. Pacak Preoperative Management of the Pheochromocytoma Patient J. Clin. Endocrinol. Metab., November 1, 2007; 92(11): 4069 - 4079. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. S. Tarant, R. G. Dacanay, B. W. Mecklenburg, S. D. Birmingham, E. Lujan, and R. Green Acute Appendicitis in a Patient with Undiagnosed Pheochromocytoma Anesth. Analg., February 1, 2006; 102(2): 642 - 643. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. O. Rynn, F. L. Hughes, and B. Faley An Emergency Department Approach to Drug Treatment of Hypertensive Urgency and Emergency Journal of Pharmacy Practice, October 1, 2005; 18(5): 363 - 376. [Abstract] [PDF]
|
|
|
|
|
|
L. Amar, A. Servais, A.-P. Gimenez-Roqueplo, F. Zinzindohoue, G. Chatellier, and P.-F. Plouin Year of Diagnosis, Features at Presentation, and Risk of Recurrence in Patients with Pheochromocytoma or Secreting Paraganglioma J. Clin. Endocrinol. Metab., April 1, 2005; 90(4): 2110 - 2116. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Favier, P.-F. Plouin, P. Corvol, and J.-M. Gasc Angiogenesis and Vascular Architecture in Pheochromocytomas : Distinctive Traits in Malignant Tumors Am. J. Pathol., October 1, 2002; 161(4): 1235 - 1246. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Q.-Y. Duh Editorial: Evolving Surgical Management for Patients with Pheochromocytoma J. Clin. Endocrinol. Metab., April 1, 2001; 86(4): 1477 - 1479. [Full Text]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
