Adrenalectomy for Familial Pheochromocytoma in the Laparoscopic Era

Abstract

Objective

To report the results of treatment of patients with familial pheochromocytomas in the laparoscopic era.

Summary Background Data

The optimal surgical management of pheochromocytomas that arise in familial neoplasia syndromes may be complicated by bilateral involvement and associated endocrinopathies.

Methods

Twenty-one patients with familial pheochromocytomas (15 with multiple endocrine neoplasia [MEN] 2A, 4 with MEN 2B, 1 each with von Hippel-Lindau and neurofibromatosis type 1) underwent adrenalectomy between December 1993 and July 2001. Clinical, biochemical, and pathologic data were obtained by retrospective review of perioperative medical records, postoperative biochemical testing, and patient questionnaire.

Results

Mean age at diagnosis was 37 ± 11 years. Twenty of the 21 patients had elevated urine catecholamines, and all had radiographic evidence of an adrenal tumor or tumors. Pheochromocytoma-related symptoms were present in 11 patients (52%). One patient with MEN 2B underwent open adrenalectomy due to previous adrenal surgery and megacolon. Laparoscopic adrenalectomy was attempted in the remaining 20 patients (9 right, 11 left, 2 bilateral). Two patients (9.1%) were converted to open adrenalectomy. Intraoperative hypertensive episodes occurred in 15 patients (71%) and were easily controlled medically. Mean operative time was 216 ± 57 minutes, mean postoperative length of stay was 3.1 ± 1.3 days, and mean tumor size was 3.1 ± 1.0 cm. Minor complications occurred in three patients (14.3%) and major complications in two patients (9.5%). During a mean follow-up of 57 months, a contralateral pheochromocytoma developed in four patients with MEN 2 (33%); three of them underwent adrenalectomy. There have been no long-term complications related to hypertension or adrenalectomy.

Conclusions

This study is the largest series of patients with familial pheochromocytoma undergoing adrenalectomy during the laparoscopic era. The results suggest that the laparoscopic approach is safe and effective for managing unilateral or bilateral adrenal medullary disease in this population.

Pheochromocytomas may occur as either sporadic or familial tumors. Although most pheochromocytomas are sporadic lesions, the familial tumors are an important group because of the presence of associated endocrinopathies and the potential for bilateral adrenal medullary disease. Familial conditions associated with the development of pheochromocytomas include the multiple endocrine neoplasia (MEN) type 2 syndromes, von Hippel-Lindau disease (VHL), and neurofibromatosis type 1 (NF-1). The MEN 2A and 2B syndromes are characterized by the presence of medullary thyroid carcinoma (MTC) in all patients. In up to 30% to 40% of patients with MEN 2A, pheochromocytomas develop; parathyroid hyperplasia develops in 30%. ¹ Patients with MEN 2B may also develop pheochromocytomas and are distinguished by the presence of multiple mucosa neuromas, ganglioneuromatosis of the gastrointestinal tract, a marfanoid-type body habitus, musculoskeletal abnormalities, and the absence of parathyroid hyperplasia. Patients with VHL may develop renal cysts, renal cell carcinomas, retinal angiomas, cerebellar hemangioblastomas, pancreatic cysts and tumors, epididymal cystadenomas, and inner ear tumors, in addition to pheochromocytomas.

The extent of adrenalectomy in patients with MEN 2A or 2B and a pheochromocytoma that involves only one adrenal has been controversial. Some groups have advocated bilateral adrenalectomy in these patients because of the high likelihood of subsequent development of pheochromocytomas ^2–4 and histopathologic studies that show adrenal medullary disease is bilateral in most instances. ^2,5 However, the approach to such patients at our institution has been to resect only the grossly involved adrenal because of the low risk of complications from an unrecognized pheochromocytoma in patients who undergo regular screening and the potential of death and complications from the addisonian state. ⁶

The most significant advance in adrenal surgery in the past decade has been the development of techniques for the removal of the adrenal gland laparoscopically. ⁷ Laparoscopic adrenalectomy has been shown to have many advantages over open adrenalectomy, including decreased pain, a faster recovery, and fewer complications. ^8–12 Pheochromocytomas, however, are potentially more challenging to remove laparoscopically for several reasons, including larger tumor size, increased vascularity, and the potential for intraoperative hemodynamic changes. Despite numerous reports of laparoscopic adrenalectomy during the past several years, ^13–15 few studies have specifically evaluated outcomes from pheochromocytoma, and there have been no large reported series of familial tumors managed in this manner.

During the past 20 years, our institution has conducted clinical, genetic, and biochemical screening of a large population of patients and kindred members with the MEN syndromes, including MEN 2A and 2B. The purpose of the present study was to review our experience in all patients with familial pheochromocytomas treated at our institution since beginning laparoscopic adrenalectomy and to evaluate the clinical presentation, diagnostic evaluation, and results of surgical treatment during the laparoscopic era.

METHODS

The study population consisted of all patients with familial pheochromocytoma undergoing surgery at the Washington University Medical Center from November 1993 to July 2001. The 21 patients with familial pheochromocytoma treated during this period included 15 patients with MEN 2A, 4 patients with MEN 2B, 1 patient with VHL, and 1 patient with NF-1. Informed consent was obtained for participation in this analysis under a protocol approved by the Washington University School of Medicine and Barnes-Jewish Hospital Human Studies Committee.

The screening protocol in the MEN 2 population consisted of at least annual assessment of clinical symptoms, measurement of urinary catecholamines and metabolites, and genetic mutational analysis. Patients with elevated catecholamine levels or a history of a prior radiographic abnormality in the adrenals were further evaluated by imaging with computed tomography (CT) or magnetic resonance imaging (MRI). Patients who had elevated urinary catecholamine levels and a grossly visible pheochromocytoma on radiographic imaging were recommended for adrenalectomy. Preparation for surgery pharmacologically was established by administration of α-receptor blockade with phenoxybenzamine. For patients with a unilateral tumor and no grossly visible disease in the contralateral adrenal, a unilateral approach to adrenalectomy was undertaken. Only patients with bilateral tumors on imaging underwent bilateral adrenalectomy. Laparoscopic adrenalectomy was carried out using a transabdominal lateral approach, and open adrenalectomy was performed using an anterior transabdominal approach.

After adrenalectomy, patients were evaluated by clinical assessment and measurement of urinary catecholamines and metabolites on an annual basis and, where clinically indicated, by radiographic imaging. Follow-up data were obtained by patient examination, telephone survey, and review of primary care physician and hospital medical records. Complications were graded according to severity. ¹⁶ All data were entered into a prospectively maintained MEN database registry. Data from a previously published series ⁶ detailing results in our kindred populations with MEN 2A and 2B and pheochromocytoma treated from 1956 to 1990 were used to compare the demographic and clinical features of two periods. Mutation analysis for mutations in the RET proto-oncogene (ret) was performed in the patients with MEN 2 as previously described. ^17,18 Results are expressed as mean ± standard deviation, and statistical analysis was performed using the Student t test and chi-square analysis.

RESULTS

The demographic characteristics of the 21 patients with familial pheochromocytoma in the present series are shown in Table 1. For comparison, the 58 patients previously described by our group who were treated during the prelaparoscopic era are shown also. In the present series, there were 19 patients with MEN 2 who belonged to 14 different kindreds. Compared with the earlier series, these patients at diagnosis were slightly older and there were significantly more men than women. Five patients (24%) in the current group had undergone adrenalectomy on the contralateral side an average of 13.2 ± 9.1 years (range 4–23) previously. In all five, the prior adrenalectomy had been performed using an open transabdominal approach with manual exploration and palpation of the unaffected side. All patients were American Society of Anesthesiology (ASA) class 2 or 3, and 10 patients (48%) had undergone previous abdominal surgery. The interval from initial thyroidectomy for MTC to surgery for pheochromocytoma in the patients with MEN 2A and 2B averaged 15.4 ± 9.5 years (range 0–34).

Table 1. PATIENT DEMOGRAPHICS

* One patient had von Hippel-Lindau disease and one patient had neurofibromatosis type 1.

Patients in the current series were much less likely to present with symptoms and signs of pheochromocytoma compared with the earlier period (Table 2). Only five patients (24%) in the current group had hypertension or reported headaches and sweating. Palpitations and tachycardia were the most common symptoms and were present in seven patients (33%). Ten patients (48%) were asymptomatic, compared with only four patients (7%) in the previous series (P < .001). Urinary catecholamines and metabolites were elevated in 20 of 21 patients (95%) (Fig. 1). In patients with MEN 2A and 2B, pheochromocytoma was diagnosed in all but three instances (15.8%) as a result of routine biochemical screening. One patient who presented with a symptomatic pheochromocytoma was found during evaluation to have concurrent MTC and was an index case to a new MEN 2A kindred. A second patient was initially diagnosed with sporadic MTC at an outside institution and had normal catecholamine levels, but a pheochromocytoma was found incidentally on CT scan. The third patient was evaluated before annual testing because of symptoms. Preoperative imaging consisted of CT in 16 patients (70%) and MRI in 11 patients (52%). Six patients (29%) had imaging with MRI to clarify findings seen on CT scans done at other institutions.

Table 2. PRE-OPERATIVE SYMPTOMS AND SIGNS

Figure 1. Results of preoperative 24-hour urinary catecholamine and metabolite testing in patients with inherited pheochromocytomas. The numbers in the bar graphs indicate the percentage of test results that were elevated. Epi, epinephrine; NE, norepinephrine; VMA, vanillylmandelic acid; combined, patients who had any one or more abnormal test results.

Perioperative data from the 21 patients in the current series are shown in Table 3. A total of 24 adrenalectomies were carried out in the 21 patients. Two patients underwent bilateral procedures, and five unilateral procedures were in patients who had had a prior open adrenalectomy. One patient with metachronous tumors also underwent a contralateral unilateral adrenalectomy at our center 5 years after the initial procedure. The surgical approach was laparoscopic in 18 patients, laparoscopic converted to open in 2 patients, and open adrenalectomy in 1 patient. One laparoscopic procedure was converted because of bleeding from the adrenal capsule and periadrenal vessels that interfered with visualization and dissection, and the other conversion was due to disruption of the adrenal capsule near the tumor. One patient with MEN 2B early in the series underwent open transabdominal adrenalectomy because of a prior open adrenalectomy and the presence of megacolon. Concomitant procedures were performed in eight patients (38%) and included liver biopsy (n = 3), adhesiolysis (n = 2), adhesiolysis with liver biopsy (n = 1), ovarian cystectomy (n = 1), and neck reoperation for MTC (n = 1). In two patients, the liver biopsy was positive for metastatic MTC.

Table 3. PERI-OPERATIVE DATA

* Includes one patient who underwent contralateral laparoscopic adrenalectomy at an interval of 63 months after laparoscopic removal of the other side, for a total of 22 separate adrenalectomy procedures in the 21 patients.

Operative time for unilateral adrenalectomy averaged 216 ± 57 minutes, and the average blood loss was 168 mL. No patient required transfusion. Hemodynamic changes of hypertension and/or tachycardia occurred during surgery in 15 patients (71%) and were controlled with intravenous administration of nitroprusside and/or nitroglycerin or labetalol. Intraoperative hypertension was mild in most instances, and only two patients experienced systolic hypertension of 200 mm Hg or more. Patients could tolerate a regular diet an average of 2.0 ± 0.8 days after surgery and were discharged home a mean of 3.1 ± 1.3 days after surgery. Four patients underwent neck reoperation for recurrent MTC within 1 week of adrenalectomy; in one patient the neck procedure was performed during the same anesthetic as the adrenalectomy.

Complications occurred in five patients (23%). Minor grade 1 complications occurred in three patients (14.3%) and consisted of fever and atelectasis, self-limited postoperative bleeding that did not require transfusion, and prolonged subcostal pain after open adrenalectomy. Grade 2 complications occurred in two patients and comprised a urethral injury from a urinary catheter and a trocar site hernia that was repaired on postoperative day 7. There were no perioperative deaths.

On pathologic examination, pheochromocytomas were present grossly and microscopically in all except one patient, who was found to have adrenal medullary hyperplasia only. Multiple tumors were present grossly in two patients. The mean size of resected tumors was 3.1 ± 1.0 cm. The results of genetic analysis for ret mutations in the patients with MEN 2 are shown in Table 4. In MEN 2A, the most common site of mutations was in codon 634 on exon 11. All four patients with MEN 2B had the same mutation involving codon 918 on exon 16. None of the patients with MEN 2B was related, and two presented as de novo mutations with no other familial involvement.

Table 4. RET PROTO-ONCOGENE MUTATION ANALYSIS IN PATIENTS WITH MEN 2A AND MEN 2B

* Mutation results are not available in two patients with MEN 2A.

† Two patients were de novo presentations of MEN 2B.

Follow-up evaluation in the entire group was carried out at a mean interval of 53 ± 27 months after surgery (range 11–95). One patient was lost to follow-up. Of the remaining 20 patients, 1 patient with MEN 2A, insulin-dependent diabetes, and mild chronic renal insufficiency has residual hypertension that is controlled medically. Of the 12 patients with MEN 2A and 2B who underwent unilateral adrenalectomy and had a remaining adrenal gland in situ, four patients (33%) developed a pheochromocytoma in the contralateral gland at a mean interval of 57 ± 8.5 months (range 44–63) after surgery (Table 5). All four of these patients had elevated urinary catecholamines and metabolites, but only one patient had symptoms possibly attributable to the tumor, and no patient was hypertensive. The tumor by CT imaging at the time of diagnosis ranged from 2 to 2.7 cm. Three of these patients underwent successful laparoscopic adrenalectomy (two at outside institutions) and the fourth patient has an asymptomatic 2-cm right adrenal mass and is being followed. No patient in this series has experienced a hypertensive crisis or other complication of a pheochromocytoma during follow-up.

Table 5. LONG-TERM FOLLOW-UP AFTER UNILATERAL ADRENALECTOMY IN PATIENTS WITH MEN 2A AND 2B

DISCUSSION

Pheochromocytomas are rare tumors that account for approximately 0.7% of cases of hypertension. Although 90% of pheochromocytomas occur as sporadic lesions, ¹⁹ inherited tumors represent an important entity because of the presence of associated endocrinopathies and the potential for early diagnosis at a minimally symptomatic or presymptomatic stage by screening at-risk individuals. Most hereditary pheochromocytomas occur in the setting of MEN 2 or VHL, ²⁰ whereas only 5% of patients with NF-1 develop pheochromocytomas. ²¹ Pheochromocytomas may occur in 30% to 50% of patients with MEN 2A and approximately 50% of those with MEN 2B. ²² In the setting of VHL, pheochromocytomas develop in 15% to 20% of patients and may be multicentric, bilateral, and extraadrenal. ²⁰ In some families, pheochromocytoma is the principal and earliest manifestation of their disease. About 44% of families with VHL (VHL type 1) do not exhibit pheochromocytomas in their disease phenotype. ²³ Pheochromocytomas associated with VHL are less likely to be symptomatic or cause hypertension than those that occur in the setting of MEN 2. ^24,25

The identification of the genetic defects responsible for MEN 2 and VHL has been a major advance in the evaluation of patients with familial neoplasia syndromes. MEN 2A and 2B arise as a result of mutations in the ret proto-oncogene on chromosomal segment 10q11, ^17,18 and VHL occurs due to mutations in the VHL tumor suppressor gene on chromosomal segment 3p25. ²⁶ In kindred members at risk for developing MEN 2, identification of individuals with germline ret mutations has allowed prophylactic total thyroidectomy to be performed to prevent the development of MTC. ^27,28 Correlation of specific ret mutations with phenotype may ultimately allow more precise identification of affected individuals who are at high risk for developing pheochromocytomas. Most of the pheochromocytomas associated with MEN 2A have involved mutations in codon 634 on exon 11 of ret and, less commonly, codon 620 on exon 10. ^29–31 Codon 634 was also the most common site mutated in our analysis, but mutations were also seen at codons 609 and 618. The most common ret mutation reported to be associated with pheochromocytomas in MEN 2B has involved codon 918 on exon 16, ²⁹ which was the only site affected in our MEN 2B patients with pheochromocytomas.

Patients with germline ret mutations for MEN 2A or 2B should undergo annual clinical and biochemical evaluation for pheochromocytoma. In this study, pheochromocytomas were identified in all but three patients (15.8%) as a result of periodic biochemical screening with 24-hour urine determination of catecholamines and metabolites. Consequently, the frequency of hypertension and symptoms attributable to pheochromocytoma in our patients was markedly decreased compared with the earlier reported series. ⁶ Recently, measurement of plasma metanephrine and normetanephrine has been shown to have high sensitivity and specificity as a diagnostic screening test for pheochromocytoma. ³² The availability of a single blood test for screening, done in conjunction with measurement of plasma calcitonin levels for MTC, has the potential to simplify screening and enhance patient compliance. Suspicion of a pheochromocytoma should precipitate radiographic imaging with CT or MRI. Our preference has been to use MRI because of the characteristic bright appearance of pheochromocytomas on T2-weighted imaging. ³³ None of the patients in this study underwent imaging with ¹³¹I-meta-iodylbenzylguanidine (MIBG) because of the limits of this modality in distinguishing adrenal medullary hyperplasia from a small pheochromocytoma.

In our institution, adrenalectomy is recommended only for patients who have elevated catecholamine levels and a grossly imageable tumor on CT or MRI. Patients with a unilateral tumor and a normal contralateral adrenal on imaging undergo unilateral adrenalectomy only. Although in the past a more aggressive approach to adrenal medullary disease has been advocated in this population, ^3,4 we have reserved bilateral adrenalectomy for patients who have gross tumor involvement of both adrenals. In the present study, the risk of developing a contralateral pheochromocytoma in patients with MEN 2 was 33% during 5 years of follow-up. In addition, in the five patients who had undergone prior open adrenalectomy, presentation with a contralateral pheochromocytoma occurred on average 13 years after the original procedure; in two patients it did not develop until 23 years later. These results compare favorably with the open adrenalectomy-era data reported by Lairmore et al ⁶, in which the incidence of development of a contralateral pheochromocytoma was 52% during 11.9 years. None of the patients in either of these two series experienced a hypertensive crisis or other complication of an undiagnosed pheochromocytoma during long-term follow-up. These results, in conjunction with the minimally invasive nature of laparoscopic adrenalectomy, argue strongly for a selective unilateral approach to patients with familial pheochromocytoma to preserve and maintain functioning adrenal tissue for as long as possible.

Despite this conservative approach, of the 21 patients in this series, 10 (48%) have no remaining adrenal tissue in situ. Although none of the patients in this study developed acute adrenal insufficiency during follow-up, the risks of total adrenalectomy must be considered in weighing surgical options. An alternative approach for selected patients may be partial adrenalectomy with preservation of adrenal cortical function. Partial adrenalectomy for pheochromocytoma in patients with MEN 2 and VHL has been reported by several groups. ^34–39 Lee et al ³⁶ carried out partial adrenalectomy using an open transabdominal approach in 14 patients with MEN 2 and VHL. Thirteen patients (93%) had normal postoperative cortisol levels and were weaned off replacement therapy. Recurrent pheochromocytomas developed in three patients (21%), and nine other patients were alive without recurrent tumor at a mean of 90 months of follow-up. Inabnet et al ⁴⁰ performed transplantation of adrenal cortical tissue into the rectus abdominis muscle in 14 patients after resection of pheochromocytomas. Although there were no recurrences of pheochromocytomas, only four patients could be weaned from exogenous steroids, and unequivocal evidence of graft function was shown in only one patient. Recently, partial adrenalectomy has been accomplished laparoscopically in a few patients. ^41–43 No tumor recurrences have been reported, but follow-up has been short and operative times have been prolonged.

Patients with familial pheochromocytomas should be an ideal group for a minimally invasive approach to adrenalectomy. Because they are usually diagnosed by screening, these patients tend to be younger, with smaller tumors and a lower incidence of hypertension, than do patients with sporadic pheochromocytomas. ⁴⁴ However, only a few series have specifically examined the results of laparoscopic adrenalectomy for pheochromocytoma, ^43,45–47 and the vast majority of reported cases have consisted of sporadic tumors. These studies show that although laparoscopic adrenalectomy for pheochromocytoma may be more difficult technically due to increased tumor size and vascularity, both solitary and multiple pheochromocytomas can be removed safely using a laparoscopic approach. A limitation of the laparoscopic approach to adrenalectomy is the inability to examine or palpate the contralateral side during surgery. However, given the sensitivity of current imaging techniques, the potential for missing a contralateral tumor should be minimal. ⁴⁸

An additional benefit of laparoscopic adrenalectomy for pheochromocytoma may be a reduction in the severity of intraoperative hemodynamic changes due to less tumor manipulation. Fernandez-Cruz et al ⁴⁹ showed that patients with pheochromocytoma treated laparoscopically had less pronounced hemodynamic changes and a lesser degree of elevation in plasma catecholamines during tumor manipulation than was seen with open adrenalectomy. Inabnet et al, ⁵⁰ however, found no differences in intraoperative hypertensive events between patients undergoing laparoscopic and open procedures. Hypertension and tachycardia may also be precipitated by the initial CO₂ insufflation (personal observation). ⁴⁶ The intraoperative hypertension that developed in 15 patients in the present series was usually mild and was easily controlled medically in every instance.

The reported conversion rate for laparoscopic adrenalectomy has ranged from 0% to 18%. 10,13–15,51–54 In the present study, two patients were converted to an open procedure. These 2 were the only conversions to open adrenalectomy in a total of 72 patients (conversion rate 3%) with functioning adrenal tumors approached laparoscopically by our group. ⁵⁴ A third patient with MEN 2B was selected for open adrenalectomy because of a history of megacolon and prior adrenalectomy via a bilateral subcostal incision. With increasing experience, the conversion rate should be minimal for tumors of this size, and prior upper abdominal surgery should not be a contraindication to a laparoscopic approach. Five patients in this series had a history of prior contralateral adrenalectomy and manual exploration of the unaffected side. Laparoscopic adhesiolysis and exposure of the adrenal gland were accomplished in all four attempted cases. Alternatively, a retroperitoneal endoscopic approach could be used in patients with extensive previous upper abdominal surgery. For the patient who has megacolon, an open insertion technique should be considered for initial access to minimize the risk of colon injury.

We have previously reported that operative times for laparoscopic removal of pheochromocytomas have been significantly longer than for adrenal cortical adenomas. ⁵⁴ The reasons for the relatively longer operative times in this series are likely due to a number of factors. In the two converted procedures, the conversions took place after much of the dissection had been completed laparoscopically. Other variables include the learning curve effect for multiple surgeons, variability in surgeon experience with adrenalectomy and advanced laparoscopic techniques, increased size and vascularity of pheochromocytomas, and the need to perform other laparoscopic procedures such as adhesiolysis. Despite these longer operative times, no adverse impact on patient outcomes or intraoperative hemodynamic instability was observed.

An important component to any laparoscopic procedure in the population of patients with familial endocrinopathies should be examination of the liver to detect lesions not seen on preoperative imaging. Tung et al ⁵⁵ reported results in 36 patients with a history of MTC and elevated plasma calcitonin levels who underwent laparoscopic examination of the liver before planned cervical reoperation for locally recurrent MTC. Metastatic MTC not seen on preoperative imaging with CT or MRI was detected in seven patients (19%), which precluded cervical reoperation in these patients. Four patients in the current series underwent liver biopsy, two of which were positive for metastatic MTC. Elevated plasma calcitonin levels were present before surgery in both patients.

Malignant pheochromocytoma is rare in the setting of MEN 2. None of the tumors in any of the MEN 2 kindreds followed up by our group have been malignant, and there have been no local or distant recurrences during long-term follow-up. However, recurrences may present up to 15 years after resection of an apparently benign pheochromocytoma. ^56,57 The importance of long-term monitoring of these patients clinically and biochemically, therefore, cannot be overemphasized. Functionally, the outcome of patients reported in this series has been equivalent to that from open adrenalectomy series for this population. Hypertension resolved completely in all except one patient due to associated medical conditions. No patient has experienced a hypertensive crisis or other untoward event related to an undiagnosed pheochromocytoma during follow-up.

In summary, we have shown that a laparoscopic approach to unilateral or bilateral adrenal medullary disease is appropriate for most familial pheochromocytomas. Patients who are diagnosed by regular clinical and biochemical screening typically have minimal if any symptoms and do not usually have marked intraoperative hemodynamic changes. For patients who have undergone unilateral adrenalectomy, the risk of developing a contralateral pheochromocytoma during follow-up is 33% during 5 years, and the risk of hypertension-related complications is low. Laparoscopic adrenalectomy has many advantages for these patients and should be the preferred surgical approach.

Acknowledgments

The authors thank Samuel A. Wells, Jr., MD, whose longstanding interest in MEN and clinical and basic investigations in patients with MEN 2 made this study possible.

Discussion

Dr. Richard E. Goldstein (Nashville, TN): It really is a nice opportunity to be able to comment on this work, where the authors clearly show that for patients with MEN syndrome and pheochromocytomas, the procedure of choice should be laparoscopic unilateral adrenalectomy.

I thought a couple of things were really very interesting. One of these deals with the fact that only 24% of your population had hypertension. I think a lot of series show that there are some patients who have pheochromocytomas who don’t have hypertension. But generally in those series, the rate is about 10%. What is different in this series compared to others that are out there, as well as the earlier group that Dr. Lairmore as well as Dr. Wells had written up back in 1993, is that in this group, the percent with hypertension is markedly lower, yet most of these people do have elevated urinary catecholamines.

My questions to you revolve around the fact that a lot of these families are now known to you. Thus, are they being screened sooner, and are you picking it up sooner? Are their levels less than in the previous series? But even with that, I think it is interesting that you may be picking up something that really could be studied at a future date in the sense that these patients do have elevated CATs probably existing for some time, and yet really most of them don’t have hypertension. So is there some phenomenon taking place where the elevation must be present for a certain period for them to really have symptoms present?

And lastly, I wanted to ask a little bit about resident teaching, because I think that has also been one of the themes of this meeting. I think all of us who do procedures like this have struggled with how to best educate house staff doing fairly difficult laparoscopic procedures, and I wondered whether you might tell us what your approach has been to trying to get these procedures done and yet educate your residents. Because, as probably a lot of people in the audience know, just a small amount of blood, maybe only a couple of tablespoons, can make it very difficult to visualize structures when it is being done laparoscopically.

Dr. Robert Udelsman (New Haven, CT): Dr. Brunt and his colleagues from the Washington University School of Medicine have had a long-term interest in familial pheochromocytoma, and their previous report published in “Annals of Surgery” in 1993 was a landmark paper in which they suggested that unilateral adrenalectomy was appropriate in these patients with unilateral tumors since it obviated the potential for life-threatening addisonian crisis. The current series supports this hypothesis, and I as well as many other surgeons have already adopted a selective approach to patients with familial pheochromocytoma. The rules for pheochromocytomas have certainly changed. Routine intraabdominal exploration with contralateral exploration of so-called normal adrenal glands is no longer the standard of care.

I did note that your length of stay in the manuscript was 3.1 days. Our practice has been to pretreat all patients with phenoxybenzamine and admit them the morning of surgery. We do not routinely use Swan-Ganz catheters, and our patients are routinely discharged on the first postoperative day on the evening following surgery. I would be curious about your perioperative protocol and how you managed these patients, particularly your anesthetic management.

Although not discussed in the presentation today, in the manuscript the authors review the world’s experience with partial or cortical-preserving adrenalectomy, an important and interesting concept based on the premise that one can preserve normal cortical function if one limits one’s dissection. I view this concept with caution. Even if their patient appears to have adequate glucocorticoid reserve after one of these procedures, it should be noted that this is under basal but not stress conditions, and I believe the surgeons may have a false sense of adrenal sufficiency when in fact the patient is adrenal insufficient. Again, Dr. Brunt, with your experience, I wonder what your thoughts are on that procedure.

Finally, I would like you to comment, if you could, about any cost comparisons comparing laparoscopic and open adrenalectomy.

Once again I congratulate the authors for their work at Barnes. It was a clearly presented presentation and a well-written manuscript, and I thank the Society for the privilege of the floor.

Dr. Murray F. Brennan (New York, NY): I run the risk here of being considered a somewhat fossilized curmudgeon, but there are two historical things I should like to comment on.

First of all, the idea of unilateral adrenalectomy for familial pheochromocytoma was the active practice at the NIH in the mid-1970s. It was really only the Mayo Clinic group who ever vigorously proposed immediate bilateral adrenalectomy based on a MEN diagnosis.

Second, laparoscopic adrenalectomy is accepted as the preferred route because it is usually compared to open adrenalectomy. There are still people in the room, of course, who grew up doing bilateral adrenalectomy for metastatic breast cancer through the posterior approach. The posterior approach is done through an incision approximately the same length as the combined incisions of the laparoscopic approach. It is done retroperitoneally. It takes about 45 minutes, rarely an hour, and the patient goes home on the first or second postoperative day.

I offer those two historical observations, and thank you for tolerating my senility.

Dr. Demetrius Pertsemlidis (New York, NY): I just wanted to make some comments on the specificity or nonspecificity of symptoms in pheochromocytoma. There are no specific symptoms to this syndrome. There is a broad spectrum of nonspecific symptoms, and dominant among these is the constellation of headaches, palpitations, and sweating. The spectrum is so wide, ranging from neurological to gastrointestinal or cardiac manifestations, including cardiomyopathy.

The other comment I would like to make is the incidence of intraoperative hypertensive crises. In the brief summary, the authors mention 15% intraoperative events. The fact is that during surgery for pheochromocytoma there are hypertensive crises or episodes of arrhythmia in almost every single one of these patients, easily controlled, because of the inevitable manipulation of the tumors; whether you do it by laparoscopy or open, compression of the tumors leads to sudden flux of catecholamines into the circulation.

Dr. Gene Branum (Harrisonburg, VA): I enjoyed the paper very much. Until I left Emory 4 months ago, we had done a number of pheochromocytomas in the past 5 years laparoscopically. The only one we encountered that we could not complete laparoscopically was a 6-cm lesion. I was wondering if there was a size limitation on pheochromocytomas that you do or do not apply to other adrenal tumors.

Second, it sounds like you are finding patients with small and asymptomatic pheochromocytomas. Might they be managed medically and not surgically in the long term?

Dr. L. Michael Brunt (St. Louis, MO): I would like to thank all the discussants for their comments.

First, in response to Dr. Goldstein. I don’t think there is any question that we are picking up some of these patients earlier than we did 10 and 20 years ago, due to increased recognition of patients with this syndrome and an aggressive approach to screening kindred members once an individual has been identified, the benefits of genetic testing and annual screening. So I think that the data from our series reflects an earlier diagnosis and detection in these patients.

I don’t have the raw catecholamine data from the previous series in order to compare. But I can tell you that in our current series in many patients really the degree of elevation was quite mild. And again I think that reflects the earlier presentation. These patients probably need to have these lesions for a period before they develop symptomatic hypertension.

You asked about our approach to teaching. This is always a challenge. We have a very broad-based experience for our residents in advanced laparoscopic techniques, and most of them by the time they reach the most senior years in the program are quite adept at doing advanced procedures. The problem with adrenalectomy is that it is an uncommon procedure, and most trainees may only get experience in one or two or three cases by the time they finish their residency. What I try to do additionally is to provide a senior resident with a videotape of the way I do the procedure before they scrub with me so they at least have some idea about the exposure and technique. And most of them are able to do this, except in the more difficult and larger tumors.

Dr. Udelsman, you asked about our length of stay. I knew this would be an issue. Three days sounds long for a laparoscopic procedure. I think there are several reasons for it. We had five patients who had a previous adrenalectomy and an additional two who had bilateral procedures. So that is seven patients who were without any remaining adrenal. Some of those were in the hospital longer in order to adjust their replacement therapy. Most of our patients were from out of state and therefore were in the hospital a little longer to some extent because of social reasons. And many others were scheduled to undergo additional procedures like neck reoperation within a few days. And we included in the length of stay the patients who had an open adrenalectomy or conversion. So I think those are the reasons for the reported length of stay.

Our perioperative protocol is similar to what you described. Patients are on phenoxybenzamine usually for 5 to 10 days preoperatively. That is usually done as an outpatient; unless an individual has a highly vasoactive tumor with a lot of marked hemodynamic changes, we usually bring them in the morning of surgery. Our intraoperative monitoring consists of an arterial line in addition to a urinary catheter and one or two good peripheral IVs.

We have not done a cost comparison in this group. We did one earlier in a reported series of laparoscopic versus open adrenalectomies in 1996. I don’t think those data probably reflect the current experience. But I believe that the laparoscopic approach is very cost-effective when you consider all the various features.

Dr. Brennan, I appreciate the historical perspective on this procedure. You commented on the posterior approach. The way we currently do laparoscopic adrenalectomy is with three 5-mm incisions and one 11-mm. I think that is less than an open posterior incision. We do have to enlarge these a little bit to get the tumor out.

In the best comparative series that I know of from the Mayo Clinic, Dr. Geoff Thompson did a case-controlled study of laparoscopic versus open posterior adrenalectomy, and the patients who had the open posterior procedure had a very high incidence, almost 50%, of long-term postoperative incisional complaints of numbness, laxity, and/or pain. So I think there is a distinct advantage to the laparoscopic approach over both types of open procedures, even if it is a bilateral procedure.

There was a comment about the hypertensive crises. I wouldn’t describe these as crises in many of those patients, since they had very mild blood pressure changes. We included any patient who had any kind of medical therapy for their blood pressure or tachycardia intraoperatively by anesthesia. As you saw from the slide I showed, those changes were minimal in most patients. But we did see some change often with tumor manipulation. And the other time you often see it is right after the initial CO₂ insufflation, before you have done anything at all with the tumor, and then they settle until you start manipulating the tumor.

Dr. Branum asked about the size limit for pheochromocytomas. The largest I have removed was 8.5 cm. I think up to 10 cm is certainly reasonable. A lot of that really depends on your confidence and experience in doing these operations. You certainly would not want to tackle one of the larger tumors early in your experience.

The question about what to do with these small tumors in patients who are asymptomatic I think is an intriguing one, and certainly one could possibly make the argument for following patients, probably at more frequent intervals, especially if it is their last remaining adrenal and they are asymptomatic, have minimal elevation in catecholamines, and no hypertension, rather than going directly to adrenalectomy.