Pulmonary Metastases of Endocrine Origin* The Role of Surgery
Junaid H. Khan, MD; Doff B. McElhinney, MD; Sarah B. Rahman, MD; Tracy I. George, MD; Orlo H. Clark, MD; and Scot H. Merrick, MD
Purpose: To determine the clinical course and outcome of patients undergoing pulmonary resection for metastatic endocrine tumors.
Methods: Retrospective review of 47 patients with known endocrine tumors and pulmonary metastases who were evaluated for surgical resection between 1975 and 1996.
Results: Tumors evaluated included the following: carcinoid (16), thyroid (12), pancreatic adenocar- cinoma (10), adrenocortical carcinoma (6), pheochromocytoma (2), and parathyroid (1). Thirty-three patients were asymptomatic. Hormone secretion was noted in five patients. Twenty-five patients, who had isolated lung metastases, good control of the primary tumor, and no medical contraindication had surgical resection. The number of pulmonary nodules was not a limiting factor as long as all disease could be resected with adequate residual pulmonary function. CT was successful in directing resection in all patients. Twenty-six operations were performed in 25 patients and 22 patients were treated medically. Wedge resection was performed for lesions <2 cm (15), and lobectomy for larger or multiple nodules (10). Four patients had bilateral nodules resected. There was no operative mortality and no major complications. Actuarial 5-year survival was 61% for surgically treated patients. Independent predictors of poor survival included positive mediastinal lymph nodes at time of surgery (p=0.004) and shorter disease-free interval (p=0.01). At a median of 6.7±1.2 years, six patients have developed radiographic appearance of a recurrence. A single patient with recurrent Hürthle cell cancer has had a successful reresection. The remaining patients have received chemotherapy. No patient with pancreatic carcinoma or adrenocortical carcinoma was a candidate for resection. All medically treated patients died within 6 months.
Conclusion: Patients with endocrine tumors and pulmonary metastases are usually asymptomatic, their conditions are diagnosed accurately with CT, and they can achieve long-term survival comparable to other tumors (sarcoma) after pulmonary metastasectomy.
Clinical implications: Patients with carcinoid, thyroid, pheochromocytoma, and parathyroid tumors with pulmonary metastases should undergo surgical resection if there is the following: (1) no evidence of extrathoracic disease; (2) good control of the primary tumor; (3) no medical contraindications for surgery; and (4) pulmonary function that can tolerate resection of all documented disease. The role of adjuvant chemotherapy in patients with positive lymph nodes needs further study.
(CHEST 1998; 114:526-534)
Key words: endocrine; metastases; pulmonary; surgery Abbreviations: MIBG=meta-iodobenzylguanidine
P rior to 1965, patients with pulmonary metastases were primarily treated with radiation therapy and/or chemotherapy. For most patients with pulmo-
*From the Division of Cardiac Surgery (Drs. Khan, Merrick, and McElhinney), University of California San Francisco, and the Departments of Surgery (Dr. Clark) and Medicine (Drs. Rah- man and George), University of California San Francisco- Mount Zion Hospital, San Francisco, CA.
Presented as a poster at CHEST 1997, the Annual Scientific Assembly of the American College of Chest Physicians, October 26-30, 1997, New Orleans.
Manuscript received September 22, 1997; revision accepted February 17, 1998.
Correspondence to: Scot H. Merrick, MD, Division of Cardiac Surgery, UCSF, 505 Parnassus, M593, San Francisco, CA 94143
nary metastases, the expected 6-month survival with this approach was approximately 50%.1-4 Although improvement in chemotherapy will undoubtedly oc- cur, substantial improvements in overall survival will likely require combined modality therapy and new methods of treatment. Currently, primary or adju- vant surgical resection of pulmonary metastases is an accepted treatment option for many patients with tumors, with an overall 5-year survival of approxi- mately 30%.2-15 Survival is highly dependent on the type of tumor, its histology and differentiation, dis- ease-free interval, number of metastases, and the presence of mediastinal nodal disease (Table 1).
| Source, yr | Cell Type | Survival | n | +/- Significant Factors |
|---|---|---|---|---|
| Edlich et al,7 1966 | Adenocarcinoma (GI, GU, breast) | 24% @ 5 yr | 61 | -Cell type |
| Sarcoma (bone, muscle) | +No. of nodules +Disease-free interval | |||
| Thomford et al,9 1965 | Adenocarcinoma (GI, GU, breast) Sarcoma (bone, muscle) | 30% @ 5 yr | 205 | -No. of nodules + Bilateral disease + Disease-free interval + Local mediastinal nodes + Extent of operation |
| Martini et al,4 1971 | Sarcoma (bone, muscle) | 45% @ 3 yr | 22 | - Disease-free interval -No. of nodules |
| Morton et al,11 1973 | Sarcoma (bone) | 60% @ 5 yr | 60 | -No. of nodules +Tumor doubling time |
| Wilkins et al,5 1978 | Adenocarcinoma (GI, GU) Melanoma | 30% @ 5 yr | 142 | + Extent of operation +Cell type |
| Morrow et al,6 1980 | Adenocarcinoma (GI, GU, breast) Melanoma Sarcoma (bone, muscle) | 29% @ 5 yr | 167 | -No. of nodules +Disease-free interval + Local mediastinal nodes |
| Wong et al,3 1993 | Melanoma | 25% @ 5 yr | 46 | +No. of nodules |
| Van Geel et al,15 1996 | Soft tissue sarcoma | 38% @ 5 yr | 255 | +Disease-free interval +Age >40 yr |
| Fourquier et al,12 1997 | Renal cell carcinoma | 44% @ 5 yr | 50 | +Complete resection |
| Robert et al, 13 1997 | Sarcoma | 40-55% @ 5 yr | 276 | - Disease-free interval |
| Teratoma | -No. of nodules | |||
| Carcinoma | ||||
| Melanoma | ||||
| Pastorino et al,14 1997 | Epithelial (n=2,260) Sarcoma (n=2,173) | 36% @ 5 yr | 5206 | + Disease-free interval |
| Germ cell (n=363) | ||||
| Melanoma (n=328) |
*Plus sign=statistically significant in predicating survival; minus sign=not statistically significant in predicting survival; GU=genitourinary.
Malignant endocrine tumors infrequently metas- tasize to the lungs and treatment options are limited. However, well-differentiated occult pulmonary me- tastases in patients with papillary and follicular thy- roid cancer can be cured with 131I therapy in about 75% of patients, whereas, 10% of patients with thyroid pulmonary metastases seen on chest radio- graphs can be treated successfully.16-18 Surgical re- section of pulmonary metastases should be consid- ered in patients with solitary or several pulmonary metastases. The surgical resection of pulmonary metastases of endocrine origin is widely accepted. However, there are little data on the outcome of this approach. The purpose of this article is to present our experience with pulmonary resection of endo- crine tumors and to examine what clinical factors may be important in predicting outcome.
MATERIALS AND METHODS
We conducted a retrospective analysis of 47 patients with known endocrine tumors who were evaluated for surgical resec- tion of pulmonary metastases between 1975 and 1996. These patients were referred to our institution for diagnosis and/or treatment. Initial diagnostic studies included serum electrolytes,
chest radiographs, and chest CT scans. Serum hormone assays and radionuclide scans were obtained in many cases to assist tumor localization and assess tumor functional status (Table 2).
All patients who were not surgical candidates were considered for chemotherapy. Twenty-two patients were treated with che- motherapy alone and multiple agents were used over the study interval. A single patient with pancreatic islet cell carcinoma received palliative radiotherapy alone.
Surgical resection was considered for patients with (1) no evidence of extrathoracic disease, (2) good control of the primary tumor, (3) no medical contraindication to pulmonary resection, and (4) all nodules resectable (no absolute upper limit on the number of nodules) with satisfactory residual pulmonary func- tion. Twenty-five patients satisfied all of these criteria and were grouped according to the histology of the primary tumor. Patients with isolated unilateral or bilateral nodules <2 cm underwent wedge resection. Lobectomy was performed for multiple lesions in the same lobe or for lesions >2 cm. Standard mediastinal lymph node dissection was performed in all patients. Patients with positive mediastinal nodes were referred for postoperative chemotherapy. The metabolic effects from the hormone-secret- ing pheochromocytoma metastases were controlled preopera- tively with a- and ß-blockade (phenoxybenzamine, esmolol).
All patients had a follow-up chest CT scan within 6 months and then yearly thereafter. Follow-up screening studies were per- formed according to tumor histology (Table 2). All patients who were identified by chest CT scans to have recurrent thoracic disease were evaluated for reresection by the same criterion used at the initial operation. One patient with a metastatic Hürthle cell
| Histology | Total No. of Patients | Patients With Resectable Disease (%) | Diagnostic Modalities |
|---|---|---|---|
| Thyroid papillary | 5 | 4 (80) | Thyroid function tests, thyroglobulin CXR, CT 1311, 201T1, 99mTc, sestamibi scans |
| Thyroid Hürthle | 2 | 2 (100) | Same |
| Thyroid medullary | 2 | 2 (100) | Calcitonin, CXR, CT |
| Thyroid anaplastic | 2 | 0 (0) | CXR |
| Thyroid lymphoid | 1 | 0 (0) | CXR |
| Carcinoid | 16 | 14 (88) | Urinary 5HIAA, CXR, CT |
| Pheochromocytoma | 2 | 2 (100) | Urinary metanephrine, CXR, CT |
| Parathyroid | 1 | 1 (100) | PTH, Ca, CXR, CT |
| Adrenal | 6 | 0 (0) | CXR |
| Pancreas | 10 | 0 (0) | CXR |
*131I=labeled iodine; 201Tl=labeled thallium; 99mTc=labeled technetium; 5HIAA=5-hydroxy indole acetic acid; CXR=chest radiograph; PTH=parathyroid hormone level; Ca=serum calcium.
carcinoma met these criteria and had reresection twice in 2 years and is asymptomatic. The remaining patients in this group were referred for multidrug chemotherapy. The chemotherapy regi- men included fluorouracil, doxorubicin, and cyclophosphamide. Streptomycin was used for patients with carcinoid tumors, and biphosphonates, diuretics, and mithramycin (plicamycin) were used for patients with parathyroid cancer.
Statistical analysis was performed (SPSS for Windows v 6.01; SPSS Inc; Chicago, IL). Fisher’s Exact Test, x2 analysis, and t test were used to compare demographic variables and disease-free intervals for surgical and nonsurgical groups. A Cox proportional hazards model was used to assess the significance of various factors as predictors of decreased postoperative survival and freedom from recurrence. Significant (p<0.05) factors were entered into multiple logistic regression according to the back- ward stepwise method.
RESULTS
There were 28 male and 19 female patients with a mean age of 56 years (12 to 90 years). Most patients (n=33) had asymptomatic pulmonary lesions at the time of presentation. Fourteen patients, however, had symptoms that included symptoms of excess hormone secretion (n=5), upper respiratory tract infections (n=4), hemoptysis (n=4), and hoarseness (n=1). The presence of symptoms did not affect postoperative survival (p=0.36).
Of the 22 patients treated medically, all had multiple bilateral pulmonary nodules and 18 had radiographic evidence of mediastinal disease. None of these patients survived for >6 months. The demographics of this group are presented in Table 3. Different chemotherapy regimens were utilized dur- ing the course of the study and although some patients developed temporary relief of symptoms and some tumor and hormonal responses were noted, these results were transient and a clear survival advantage was not evident. The role of
chemotherapy and radiation therapy in these pa- tients with clearly more advanced disease needs further study.
The characteristics of 25 patients undergoing sur- gical resection are presented in Table 3. In this group, chest CT scanning was the most accurate method for tumor localization as it correctly identi- fied the number and location of pulmonary nodules in 85% of cases. Additional nodules discovered at surgery did not affect resectability. The results of other imaging techniques did not influence surgical consideration or operative technique. However, a meta-iodobenzylguanidine (MIBG) scan did help confirm the diagnosis of pheochromocytoma in two patients. The most common surgical approach was thoracotomy in 21 patients, with median sternotomy being used in 4 patients with bilateral lesions. Wedge resection was performed in 15 of 25 patients (60%), while the other 10 patients underwent a lobectomy. None of the surgical patients had any radiographic
| Surgical | Nonsurgical | |
|---|---|---|
| N | 25 | 22 |
| Age, yr, mean±SD | 46±17 | 63±23 |
| Disease-free interval (median) | 9 mo (6 mo-8 yr) | N/A |
| Single lesion, No. | 13 | 0 |
| Unilateral: multiple, No. | 8 | 0 |
| Bilateral: multiple, No. | 4 | 22 |
| Surgical approach, No. | Thoracotomy, 21 Sternotomy, 4 | |
| Extent of resection, No. | Wedge resection, 15 Lobectomy, 10 | |
| Actuarial survival | 6 mo, 85% | 6 mo, 0% |
| 1 yr, 79% | ||
| 5 yr, 61% |
evidence of mediastinal disease preoperatively. Me- diastinoscopy was not performed in any patient. At the time of surgery, mediastinal lymph node sam- pling was performed.
There were no operative deaths. One patient had a prolonged air leak (5 days), which was treated with a tube thoracostomy, and one patient experienced transient atrial fibrillation.
During a mean±SD follow-up of 6.7±1.2 years, six patients have developed radiographic evidence of recurrent disease. In five patients, the recurrence was in the mediastinal lymph nodes, while a single patient had parenchymal recurrence. The median interval from surgery to diagnosis of mediastinal disease was 6 months (4 to 9 months). Two of these patients had positive mediastinal nodes at the time of pulmonary resection and had received postoperative adjuvant chemotherapy. The single patient with a parenchymal lung recurrence had a Hürthle cell carcinoma and was treated with reresection. Ten of the surgically treated patients died during the fol- low-up period, seven related to widespread metasta- sis of the primary tumor, and three to unrelated causes.
Factors associated with shorter survival are listed in Table 4. Independent predictors of shorter long- term survival by multivariate analysis included posi- tive mediastinal lymph nodes at the time of pulmo- nary resection (p=0.004) and a shorter disease-free interval (p=0.01) (Table 4). No statistical differences in survival were found between different tumor types, although this analysis was limited by the small sample size in the various diagnostic groups.
| Variable | p Value: Univariable | p Value: Multivariable |
|---|---|---|
| Age | 0.08 | |
| Female sex | 0.01 | 0.2 |
| Multiple (vs single) lesion | 0.95 | |
| Unilateral (vs bilateral) lesion | 0.72 | |
| Wedge (vs lobectomy) | 0.36 | |
| Histology of tumor: | 0.62 | |
| Thyroid: papillary/medullary/Hürthle, parathyroid, carcinoid, pheochromocytoma | ||
| Presence of symptoms | 0.36 | |
| Longer disease-free interval | 0.056 | |
| Positive mediastinal lymph nodes at surgery | 0.02 | 0.004 |
| Shorter postoperative freedom from recurrence | 0.04 | 0.01 |
DISCUSSION
Many endocrine tumors share a common embry- ologic origin and have been classified according to a common scheme amine precursor uptake and decar- boxylation. However, within this classification sys- tem, the biological behavior of individual tumors is quite variable. Among our patients, carcinoid was the most common endocrine metastasis and was symp- tomatic in 50% of patients. Four patients had symp- toms of upper respiratory tract infection and four patients had hemoptysis. No patients had carcinoid syndrome. However, patients with pulmonary carci- noid tumors can have carcinoid syndrome without other metastases, as the liver does not detoxify the pulmonary venous return. Bronchial carcinoid me- tastases also accounted for 50% of patients with multiple lesions (n=6). The diagnosis of the metas- tases occurred synchronously with the diagnosis of the primary bronchial carcinoid in all cases. Eighty percent of bronchial carcinoid metastases were re- sectable and the 5-year survival was 72%.
Carcinoids are the most common GI neuroendo- crine tumors. They are neoplasms of enterochromaf- fin (Kultschitsky) cells. The site and size of the primary tumor and presence of symptoms affect their metastatic potential. Bronchial carcinoids are less likely to metastasize than tumors of the midgut; however, they are more likely to metastasize to the lungs19 (Fig 1).
Adrenal tumors were usually suspected because of excess hormone secretion or incidentally on abdom- inal CT scans (pheochromocytoma, n=2; adrenal cortical carcinoma, n=6). Pulmonary metastases to the adrenal gland are much more common than primary adrenal tumors. Adrenal tumors with metas- tases to the lung are even more uncommon. CT scans are, in general, the imaging study of choice for localization of primary or metastatic adrenal tumors. However, MRI scanning with T1 and T2 imaging with gadolinium is preferable for patients with bron- chial carcinoids. The diagnosis is confirmed by se- rum assays for excess hormone secretion. Overall, metastases are uncommon in patients with Cushing’s disease and rare in patients with hyperaldosteronism. However, more than half of patients with adrenocor- tical carcinoma develop metastatic disease, most commonly in the lung and liver.20,21 Survival in patients with adrenocortical carcinoma confined to the adrenal gland at resection is about 30% at 5 years, but the median survival of patients with metastatic disease is 6 months.20,21 About 35% of patients respond to treatment with mitomycin.22 None of our patients with adrenocortical carcinoma and pulmonary metastases were candidates for pul- monary resection and all died within 6 months.
0
3
Most patients with pheochromocytomas present with symptoms. MRI is preferable to CT for tumor localization when T1 and T2 gadolinium-enhanced images are used, with tumor enhancement showing up on T2 imaging. MIBG scans are most sensitive for identifying extra-adrenal pheochromocytomas. Pul- monary metastases are uncommon in patients with malignant pheochromocytomas. Isolated metastases should be resected when this is technically possible with a 5-year survival approaching 60%. Patients with unresectable disease can have their symptoms palliated with @- and ß-blockade. MIBG-131I in high doses has been used in selected patients for tumor ablation and hormone control, but no clear survival advantage has yet to be demonstrated.23,24 Median survival in patients with metastatic disease remains 6 to 12 months. Both of our patients with pheochro- mocytomas had symptoms of excess catecholamine secretion after removal of the primary adrenal tu- mor. MIBG scanning helped locate the metastases and these patients are alive and disease free 57 and 62 months following resection (Fig 2).
Parathyroid tumors are usually benign adenomas that secrete excess parathyroid hormone with result- ant hypercalcemia. Ultrasound, CT, sestamibi, and MRI scanning are effective for tumor localization but they are expensive and are not necessary prior to initial parathyroid exploration because experienced surgeons can successfully treat >95% of patients without preoperative localization studies. Localiza- tion studies are most useful for patients with persis- tent or recurrent hyperparathyroidism.25,26 When results of these studies are normal, equivocal, or
conflicting, we use highly selective venous catheter- ization and parathyroid hormone assay. Parathyroid carcinoma is a rare entity (0.5% of all patients with primary hyperparathyroidism) but it is a cause of persistent or recurrent hyperparathyroidism.25-27 Most patients present with symptoms and marked hypercalcemia, which is often difficult to control medically. Metastatic disease is uncommon, with pulmonary, bone, and liver metastases being the most frequent sites. Aggressive surgical resection of distant metastases as well as local recurrences has resulted in some long-term survivors.27,28 Biphos- phonate and diuretic therapy with furosemide is used in patients with unresectable tumors; however, the prognosis remains poor. Our single patient had evidence of hypercalcemia after parathyroidectomy for an “adenoma,” and three normal parathyroid glands were identified. Her pulmonary metastasis was identified on a chest radiograph during a fol- low-up for persistent hypercalcemia. The hypercal- cemia resolved after wedge resection of the isolated pulmonary metastasis (Fig 3).
Thyroid cancers vary greatly in their growth rate and invasive behavior. Thyroid neoplasms are usually diagnosed by fine-needle aspiration cytology. Papil- lary cell tumors account for 84%, follicular and Hürthle cell 10%, medullary 5%, and anaplastic 1% of all thyroid tumors. Papillary tumors commonly metastasize to regional lymph nodes, and about 15% of patients have or develop pulmonary or bone metastasis.29 Follicular and Hürthle cell carcinomas more frequently spread to lungs and skeleton, whereas medullary thyroid cancers frequently metas-
tasize to the liver. Treatment is usually by total thyroidectomy with postoperative ablative treatment with 131I and is associated with an 80% 10-year survival.30-32
Thyroglobulin is a sensitive marker for recurrent and metastatic thyroid cancers of follicular cell origin after total thyroidectomy. Patients with micrometas- tases in lungs and bones can often be treated suc-
cessfully with 131I, whereas larger isolated metastases are better treated with resection when possible.16,18
Among our patients with papillary and follicular thyroid cancer, eight of nine (88%) of the pulmonary metastases took up 131I. Unfortunately, it did not ablate these discrete lesions. Our patients with thy- roid cancers were asymptomatic, and when metasta- ses were resectable, the patients had a good progno-
&
L
R
140 KV 100 MA LRG SFOV 10 0 MM
0 TILT
280
3 0 SEC 08:17 :27
sis (actuarial 5-year survival, 58%). The single reresection was for a patient with Hürthle cell carcinoma. She is still alive and asymptomatic, but again she has three pulmonary metastases (Fig 4).
Medullary carcinoma accounts for approximately 5% of all thyroid tumors and may be sporadic (75%) or familial (25%). Treatment, which is by total thyroidectomy and central neck dissection, has a 10 year survival of about 70%.33 Calcitonin and carci- noembryonic antigen are sensitive markers for recur- rent or persistent medullary cancer. Regional lymph node metastases occur in about 50% of patients. Metastases to skeleton, lung, and liver occur in about 10% of patients.34 Radioactive iodine has a limited, if any, role in treatment of medullary tumors since most of these cancers do not concentrate iodine. Labeled carcinoembryonic antigen has been used recently with limited success.35 Aggressive surgical resection of metastatic medullary thyroid cancer,
FIGURE 4. Top: left lower lobe metastases in patient after total thyroidectomy for Hürthle cell carcinoma. Bottom: Histology shows a well-circumscribed tumor nodule adjacent to uninvolved lung parenchyma (right upper corner). The pattern of growth is solid/trabecular with arrows indicating three follicles containing inspissated colloid. Tumor cells are pleomorphic with abundant granular cytoplasm and prominent nucleoli (original magnifica- tion, ×250).
when technically possible, has resulted in some long-term survivors as occurred in our patients, one who survived 10 months and another who is alive and disease free at 14 years.
Anaplastic thyroid cancer and thyroid lymphoma are rare tumors with a median survival of 6 month even without metastases at presentation.36,37 Neither of our two patients with anaplastic cancer nor one patient with stage IV lymphoma survived for 6 months.
Cancer of the pancreas is the third leading cause of death in the United States after lung and colon in men 35 to 54 years of age and adenocarcinoma accounts for 80% of these cases. Pulmonary and peritoneal metastases occur late in the disease. Me- dian survival for patients whose tumors are unresect- able is 6 months.38 No patient with metastatic pan- creatic carcinoma was found to have resectable disease (adenocarcinoma, n=7; islet cell tumor, n=3), and all died within 6 months.
In conclusion, pulmonary metastases are uncom-
mon in patients with malignant endocrine neoplasm. Most patients with pulmonary metastases are asymp- tomatic and the overall median survival is 6 months.1,6,7 However, this study was not designed to compare medical vs surgical treatment, since pa- tients treated medically were those not selected for surgery and these patients were expected to have a worse prognosis.
Some patients with malignant endocrine tumors, however, have slowly growing metastases, including patients with malignant carcinoid tumors, well-dif- ferentiated thyroid cancers, and parathyroid cancers. These patients appear to benefit from complete or partial resection of the metastases. 19,28,29
There is some controversy regarding resection of metastases that must be noted. Some believe the presence of metastases to imply systemic dissemina- tion of disease, and recommend surgery only for palliation of local symptoms.39 The rate of survival after resection had not changed since the 1940s, indicating the limitation of the operation and the inability of the chemotherapeutic agents to control these tumors once metastases have developed.40 However, metastatectomies can be performed with low morbidity and they can be justified by long-term disease-free interval and cases of apparent cure.
Pulmonary metastases should not preclude surgi- cal resection in patients with isolated pulmonary metastases limited to a single lobe due to malignant endocrine neoplasm. The number of pulmonary nodules does not appear to adversely influence survival as long as residual lung function after resec- tion is adequate. More than 20 nodules have been successfully resected in some series.4,41 Surgical approach utilized in this study was a posterolateral thoracotomy for unilateral lesions and a median sternotomy for bilateral lesions. Various other ap- proaches have also been described for bilateral metastases resection. Thoracoscopic resection for unilateral and bilateral lesions deserves special com- ment. At this time, we have not utilized this tech- nique in our practice due to the inability to palpate for other occult lesions. However, with the improv- ing sensitivity of CT scans, minimizing the probabil- ity of occult lesions, we are reevaluating the role of thoracoscopic resection of metastases.
We performed wedge resections as our primary mode of metastases resection. Even in cases of multiple lesions in the same lobe, multiple wedge resections were performed. Lobectomy was per- formed only if all lesions could not be technically resected, while preserving some functioning portion of a lobe. Conservation of lung parenchyma was an important factor in all metastases resection.
Patients without mediastinal nodal involvement and those with a longer duration before development
of pulmonary metastases had the best prognosis (p<0.01). The limitations of the retrospective nature of the study preclude any definitive statement about the management of mediastinal nodes. We sample mediastinal nodes for all metastasectomies. It is unclear whether endocrine tumors have a greater propensity for mediastinal nodal involvement. In our study, five of six recurrences were in the mediasti- num, but only two had had positive nodes at the time of surgery. However, further study is needed to determine whether those patients with positive nodes benefited from surgical resection.
Patients with carcinoid, pheochromocytoma, para- thyroid, and localized thyroid metastases should proceed to resection if the primary tumor is con- trolled, there is no evidence of extrathoracic disease, and there is no medical contraindication to pulmo- nary resection. This treatment not only decreases some metabolic problems such as hypercalcemia and hypertension but also appears to improve survival. Our overall 5-year actuarial survival of 61% in our patients with surgically resectable disease compares favorably with the reported survival rates with pul- monary metastases of other tumors (colon, sarcoma, melanoma, renal cell).5-7,12,14,15 The results of 10- year survival may be more appropriate in assessing surgical treatment in these favorable tumors and a follow-up study is planned. Postoperatively, chemo- therapy should be considered in patients with evi- dence of mediastinal nodal disease at the time of resection and in those patients who have radio- graphic or surgical evidence of mediastinal or diffuse pulmonary recurrence.42 Patients with isolated pul- monary recurrence should be considered for rere- section by the same criteria used initially. The benefit of adjuvant therapy in patients with positive nodes needs further study.
REFERENCES
1 Farrell JT. Pulmonary metastasis: a pathologic, clinical, roent- genologic study based on 78 cases seen at necropsy. Radiology 1935; 24:444-51
2 Takita H, Edgerton F, Karakousis C, et al. Surgical manage- ment of metastases to the lung. Surg Gynecol Obstet 1981; 152:191-94
3 Wong JH, Skinner KA, Kim KA, et al. The role of surgery in the treatment of nonregionally recurrent melanoma. Surgery 1993; 113:389-94
4 Martini N, Huvos AG, Mik V, et al. Multiple pulmonary resections in the treatment of osteogenic sarcoma. Ann Thorac Surg 1971; 12:271-80
5 Wilkins EW Jr, Head JM, Burke JF. Pulmonary resection for metastatic neoplasms in the lung: experience at the Massa- chusetts General Hospital. Am J Surg 1978; 135:480-83
6 Morrow CE, Vassilopoulos PP, Grage TB. Surgical resection for metastatic neoplasms of the lung: experience at the University of Minnesota Hospitals. Cancer 1980; 45:2981-85
7 Edlich RF, Shea MA, Foker JE, et al. A review of 26 years’
experience with pulmonary resection for metastatic cancer. Dis Chest 1966; 49:587-94
8 Mountain CF, Khalil KG, Hermes KE, et al. The contribution of surgery to the management of carcinomatous pulmonary metastases. Cancer 1978; 41:833-40
9 Thomford NR, Woolner LB, Clagett OT. The surgical treat- ment of metastatic tumors in the lungs. J Thorac Cardiovasc Surg 1965; 49:357-63
10 Damron TA, Pritchard DJ. Current combined treatment of high-grade osteosarcomas. Oncology 1995; 9:327-50
11 Morton DL, Joseph WL, Ketcham AS, et al. Surgical resec- tion and adjunctive immunotherapy for selected patients with multiple pulmonary metastases. Ann Surg 1973; 178:360-66
12 Fourquier P, Regnard JF, Rea S, et al. Lung metastases of renal cell carcinoma: results of surgical resection. Eur J Car- diothorac Surg 1997; 11:17-21
13 Robert JH, Ambrogi V, Mermillod B, et al. Factors influenc- ing long-term survival after lung metastasectomy. Ann Thorac Surg 1997; 63:777-84
14 Pastorino U, Buyse M, Friedel G, et al. Long-term results of lung metastasectomy: prognostic analyses based on 5,206 cases: the International Registry of Lung Metastases. J Thorac Cardiovasc Surg 1997; 113:37-49
15 Van Geel AN, Pastorino U, Jauch KW, et al. Surgical treatment of lung metastases: the European Organization for Research and Treatment of Cancer-Soft Tissue and Bone Sarcoma Group study of 255 patients. Cancer 1996; 77: 675-82
16 Casara D, Rubello D, Saladini G, et al. Different features of pulmonary metastases in differentiated thyroid cancer: natu- ral history and multivariate statistical analysis of prognostic variables. J Nucl Med 1993; 34:1626-31
17 Schlumberger M, Challeton C, De Vathaire F, et al. Treat- ment of distant metastases of differentiated thyroid carci- noma. J Endocrinol Invest 1995; 18:170-72
18 Schlumberger M, Challeton C, De Vathaire F, et al. Radio- active iodine treatment and external radiotherapy for lung and bone metastases from thyroid carcinoma. J Nucl Med 1996; 37:598-605
19 Moffat FL Jr, Ketcham AS. Metastatic proclivities and pat- terns among APUD cell neoplasms. Semin Surg Oncol 1993; 9:443-52
20 Icard P, Chapuis Y, Andreassian B, et al. Adrenocortical carcinoma in surgically treated patients: a retrospective study on 156 cases by the French Association of Endocrine Surgery. Surgery 1992; 112:972-80
21 Icard P, Louvel A, Chapuis Y. Survival rates and prognostic factors in adrenocortical carcinoma. World J Surg 1992; 16:753-58
22 Wooten MD, King DK. Adrenocortical carcinoma: epidemi- ology and treatment with mitotane and a review of the literature. Cancer 1993; 72:3145-55
23 Krempf M, Lumbroso J, Brendel A, et al. Treatment of malignant pheochromocytoma with [13]]] metaiodobenzyl- guanidine: a French Multicenter Study. J Nucl Biol Med 1991; 35:284-87
24 Colombo I, Lomuscio G, Vignati A, et al. Preliminary results of [13]]] metaiodobenzylguanidine treatment in metastatic malig- nant pheochromocytoma. J Nucl Biol Med 1991; 35:300-04
25 Rodriquez JM, Tezelman S, Siperstein AE, et al. Localization procedures in patients with persistent or recurrent hyperpara- thyroidism. Arch Surg 1994; 129:870-75
26 Shen W, Duren M, Morita E, et al. Reoperation for persistent or recurrent primary hyperparathyroidism. Arch Surg 1996; 131:861-69
27 Kay S, Hume DM. Carcinoma of the parathyroid gland: how reliable are the clinical and histological features? Arch Pathol 1973; 96:316-19
28 Obara T, Okamoto T, Ito Y, et al. Surgical and medical management of patients with pulmonary metastasis from parathyroid carcinoma. Surgery 1993; 114:1040-49
29 Harness JK, Thompson NW, McLeod MK, et al. Differenti- ated thyroid carcinoma in children and adolescents. World J Surg 1992; 16:547-54
30 Staunton MD, Bourne H. Malignant thyroid tumours 1932 to 1972: the outcome in 492 patients. Eur J Surg Oncol 1992; 18:469-74
31 Donohue JH, Goldfien SD, Miller TR, et al. Do the prog- noses of papillary and follicular thyroid carcinomas differ? Am J Surg 1984; 148:168-73
32 DeGroot LJ, Kaplan EL, Shukla MS, et al. Morbidity and mortality in follicular thyroid cancer. J Clin Endocrinol Metab 1995; 80:2946-53
33 Dottorini ME, Assi A, Sironi M, et al. Multivariate analysis of patients with medullary thyroid carcinoma: prognostic signif- icance and impact on treatment of clinical and pathologic variables. Cancer 1996; 77:1556-65
34 Russell CF, Van Heerden JA, Sizemore GW, et al. The surgical management of medullary thyroid carcinoma. Ann Surg 1983; 197:42-48
35 Juweid M, Sharkey RM, Behr T, et al. Radioimmunotherapy of medullary thyroid cancer with iodine-131-labeled anti- CEA antibodies. J Nucl Med 1996; 37:905-11
36 Tan RK, Finley RK, Driscoll D, et al. Anaplastic carcinoma of the thyroid: a 24-year experience. Head Neck 1995; 17:41-48
37 Kobayashi T, Asakawa H, Umeshita K, et al. Treatment of 37 patients with anaplastic carcinoma of the thyroid. Head Neck 1996; 18:36-41
38 Wade TP, Kraybill WG, Virgo KS, et al. Pancreatic cancer treatment in the US veteran from 1987 to 1991: effect of tumor stage on survival. J Surg Oncol 1995; 58:104-11
39 Barr LC, Skene AJ, Thomas JM. Metastasectomy. Br J Surg 1992; 79:1268
40 Moores DWO. Pulmonary metastases revisited [editorial]. Ann Thorac Surg 1991; 52:178-79
41 Girard P, Baldeyrou P, Le Chevalier T, et al. Surgical resection of pulmonary metastases: up to what number? Am J Respir Crit Care Med 1994; 149(2 pt 1):469-76
42 Patwarden N, Edmiston K. Chemotherapy for unresectable endocrine neoplasms. In: Clark OH, Duh QY, eds. Textbook of endocrine surgery. Philadelphia: WB Saunders, 1997; 1-16