A Contemporary Perspective on Superior Vena Cava Syndrome John C. Chert, MD, Fred Bongard, MD, Stanley R. Klein, MD, Torrance,California The superior vena cava (SVC) syndrome is usually associated with advanced malignancy and has a dismal prognosis. In order to analyze the impact o f newer diagnostic and therapeutic modalities, we retrospectively examined the last 45 consecutive cases of SVC syndrome treated over a 12-year period. The underlying causes were advanced lung cancer ( 5 7 % ) , mediastinal tumors ( 2 0 % ) , and metastatic solid malignancy ( 5 % ) . Forty-two patients ( 9 3 % ) were treated with external beam radiotherapy and/or chemotherapy with a mean patient survival of 3 months; 11 of 42 patients (26%) were treated without histologic diagnosis. Symptoms of SVC obstruction resolved in 80% of patients who underwent radiotherapy, with a mean interval of 4 weeks. The most common cause of death was respiratory arrest. Of the three patients with benign disease, two underwent eaval reconstruction with greater than 3-year potency and relief of symptoms. We conclude that ( 1 ) SVC syndrome portends a grim prognosis when associated with malignancy but usually responds to radiation or chemotherapy; (2) CT scan is the best available method to document the extent and location of involvement; and (3) patients with benign disease should be evaluated for caval reconstruction, which may produce rewarding long-term results.
uperior vena cava (SVC) obstruction, although rare, can be responsible for life-threatening laryngeal edema and cerebral congestion. Serious sequelae of respiratory distress, seizures, and intracranial venous thrombosis have been associated with atypical presentations of SVC syndrome [1]. In the majority of cases, however, slow progressive occlusion of the SVC produces mild symptoms of headache, blurry vision, and neck and facial swelling. Collateral circulation generally enlarges ade-
S
From the Department of Surgery, Harbor-UCLA Medical Center, UCLA Schoolof Medicine,Torrance,California. Requestsfor reprintsshouldbe addressedto StanleyR. Klein,MD, Department of Surgery, Harbor-UCLA Medical Center, UCLA Schoolof Medicine,1000WestCarsonStreet, Box 15, Torrance,California 90509. Presented at the 18th AnnualMeetingof the Societyfor Clinical VascularSurgery,Palm Desert,California,March7-11, 1990.
quately to alleviate any transient disabilities due to acute vena caval insufficiency. Since Hunter's [2] first authentic description of SVC syndrome due to a luetic saccular aortic aneurysm, the spectrum of causes and diagnostic modalities has evolved drastically. Prior to 1949, only one third of the SVC occlusions reported were caused by primary chest malignancies. Mclntire and Sykes [3] reported their series of 502 cases in which 67% of patients with SVC syndrome had a benign cause and only 33% had underlying malignancies. This contrasts with Fincher's [4] 1987 experience in which 87% of patients with SVC syndrome had a primary intrathoracic cancer. In this retrospective study, we attempt to identify the frequent presenting findings and common causes of SVC syndrome and the useful diagnostic modalities and therapeutic outcome of various treatment regimens at our institution over the past decade, MATERIAL AND METHODS Between November 1976 and December 1988, 29 men and 16 women were treated for SVC syndrome at the Harbor-UCLA Medical Center. Their ages ranged from 18 to 76 years with a mean age of 54 years. This includes those admitted for a different diagnosis who were found subsequently to have stigmata of SVC syndrome. The initial symptoms consisted of dyspnea (71%), headache (11%), dysphagia (9%), cough (4%), and visual disturbance (4%). Physical findings at time of presentation are listed in Table I. More than half of the patients had facial engorgement, upper extremity edema, and shortness of breath on exertion. Less common signs were jugular venous distension, plethora, upper torso swelling, and unclear vision. The underlying causes of SVC syndrome in this series are listed in Table II. SVC syndrome was caused by malignant diseases in 42 of 45 patients (93%). Primary lung cancers were the leading cause, with squamous cell carcinoma accounting for 24% of the total. Metastatic breast carcinoma (11%) was the second most common cause, while lymphoma (9%) and seminoma (6%) were less common. Unusual malignant lesions that produced SVC syndrome were metastatic colon and renal cell carcinomas. Three patients had benign diseases (7%). One of these was a man who presented with a 2-year history of facial swelling, plethora, and upper-extremity edema. After an extensive work-up, he was found to have a fibrous mediastinal mass at sternotomy. Pathologic examination showed evidence of mediastinal fibrosis without signs of malignancy. The patient was given a final diagnosis of idiopathic mediastinal fibrosis. Another patient was a woman with a history of tuberculosis who presented with 4 weeks of cyanosis and facial edema. After an inconclu-
THE AMERICAN JOURNAL OF SURGERY VOLUME160 AUGUST 1990 207
CHEN ET AL
TABLE I Physical Findings at Presentation In 45 Patients Patients Findings
n
%
Facial/neck edema Extremity edema Dyspnea on exertion Jugular venous distension Plethora Chest/shoulder swelling Blurry vision
35 34 32 12 6 3 2
78 75 71 27 13 7 4
TABLE l I Cause of SVC Obstruction In 45 Patients
Patients Cause
n
%
Lung carcinoma Squamous cell carcinoma Undifferentiated carcinoma Adenocarclnoma Oat cell carcinoma Breast carcinoma Lymphoma Seminoma Malignant thymoma Metastatic colon carcinoma Metastatic renal cell carcinoma Benign idiopathic medlastlnal fibrosis Tuberculosis lymphangitis Epicardial pacemaker wire
11 7 5 3 5 4 3 2 1 1 1 1 1
24 16 11 7 11 9 6 4 2 2 2 2 2
TABLE I I I
Diagnostic Modalltles Used In the Evaluation of SVC Syndrome Procedure
No. Done
Chest computerized tomography Superior venacavogram Thoracotomy/sternotomy Bronchoscopy Barium swallow Needle aspiration Radionuclide scan* Magnetic resonance imaging Medlastinoscopy Upper endoscopy Clinical diagnosis
23 12 6 6 4 2 2 2 1 1 12
* Technetium-99m.
sivr search, she was presumed to have tuberculosis lymphadenitis. The third patient complained of dyspnea on exertion for 6 weeks and right-sided facial edema lasting 8 months. He underwent a superior venacavogram and was found to have a small area of obstruction (narrowing) that ultimately proved to be an improperly placed epicardial pacemaker wire. Fifty-nine diagnostic procedures were performed on 33 patients to define the cause and location of their obstruction. Eleven of 45 patients underwent treatment 208
THE AMERICAN JOURNAL OF SURGERY
based on history and clinical findings alone. The procedures used are listed in Table III. The diagnostic modality most frequently employed was chest computed tomography (CT). CT scanning confirmed SVC occlusion in all 23 patients in whom it was performed. The criteria for a positive CT scan included poor opaciflcation of the SVC with intravenous contrast injection, evidence of a mediastinal mass extending into the SVC, and numerous venous collaterals demonstrated on the lateral and anterior chest wall (Figure 1). Venacavography was used in 12 patients. Superior venacavogram accurately defined the site of obstruction in all those studied (Figure 2). There were no reported complications from this procedure. Thoracotomy or sternotomy was necessary in 6 of the 45 patients (13%) with SVC syndrome, Surgery was performed solely for the purpose of obtaining tissue for diagnosis in four of the patients. The remaining two patients underwent SVC bypass grafting and excisional biopsy at the same operation. Bronchoscopy was useful in diagnosis 67% of the time; four of the six bronchoscopies performed yielded positive cytology for malignancy. Radionuclide scanning and magnetic resonance imaging (MRI) were infrequently used, but both were effective in confirming SVC obstruction. MRI showed a mediastinal mass compressing the SVC in the single patient in whom it was performed (Figure 3). Of the 42 patients with malignant disease, 33 had relief of symptoms with radiation and/or chemotherapy, 7 were lost to follow-up, and 2 died prior to initiation of therapy. Of the three patients with benign diseases, two underwent SVC reconstruction with immediate resolution of symptoms. Autologous saphenous vein grafts were used as bypass conduits in both patients. These two patients had no evidence of SVC syndrome at 3-year followup. In the third patient who sustained SVC thrombosis due to a cardiac pacing wire, anticoagulation was sufficient in providing long-term symptomatic relief. However, he died 2 years later of progressive cardiomyopathy. COMMENTS SVC syndrome is a rare clinical entity first described by William Hunter in 1757 [2]. It has a distinctive presentation that can be easily recognized once the diagnosis is entertained. Parish and co-workers [5] maintain that physical examination is all that is required in most patients. Occlusion of the venous outflow from the head, neck, and upper extremities produces characteristic findings of facial edema, cyanosis, dyspnea, and prominent neck veins that hallmark the syndrome. Obstruction can be due to thrombosis, extrinsic mass effect, or invasion of venous intima by neoplasm. Recently, an important change has occurred in the common causes of SVC syndrome. A 1949 review by McIntire and Sykes [3] showed that of 502 cases of SVC syndrome analyzed, 38% were caused by primary thoracic tumors, 30% by aortic aneurysm, 15% by chronic mediastinitis, 8% by phlebitis, 6% by metastatic disease, and 3% by an uncertain cause. Kamiya and colleagues [6] studying 734 patients from 1949 to 1965 demonstrated that 540 (74%) of them had intrathoracic malignancy as the cause of SVC syndrome.
VOLUME 160
AUGUST 1990
SUPERIOR VENA CAVASYNDROME
In recent surveys by Fincher [4] and Schraufnagel et al [7], malignant diseases accounted for 87% and 97% of the cases, respectively. Our series found a malignant cause in 93% of the cases. This trend toward malignancy is most likely a result of improved detection and treatment of granulomatous and infectious mediastinal lesions before they progress to obstruction. Benign causes of SVC syndrome have been relatively uncommon. With the declining cases of tuberculosis and syphilis, curable SVC obstructions are infrequent. Mitchell and co-workers [8] reported their experience using autogenous saphenous vein grafts to perform successful bypasses in two patients with benign idiopathic mediastinal fibrosis. Gladstone et al [9] described effective bypass in one patient with mediastinal fibrosis using the femoral vein as a conduit. Reddy et al [10], Davis and Shevland [11], and Roels et al [12] have each reported two cases of SVC syndrome due to benign goiters. Three new cases of benign eaval obstruction are discussed in the present study. One patient developed obstruction of the SVC due to idiopathic mediastinal fibrosis, a rare fibrosclerotic condition of unknown cause. Various factors have been implicated in its pathogenesis, in particular infections with tuberculosis and histoplasmosis, the drug methysergide, and autoimmune processes [13]. To date, however, absolute cause and effect has not been demonstrated with any single agent. Another patient was diagnosed as having tuberculosis lymphadenitis. Tuberculosis infection evokes a reactive cicatricial contraction in the mediastinum, with gradual obstruction of the SVC [3]. The third patient was found to have SVC syndrome following an epieardial pacemaker wire insertion. With the marked decrease in infectious lesions of the mediastinum and the recent increased use of invasive monitors such as SwanGanz catheters, pacemaker electrodes, and central alimentation catheters, iatrogenic causes may emerge as a common cause of nonmalignant SVC syndrome in the future. SVC syndrome often causes tracheal compression and airway compromise in the pediatric population. Lymphoma is the most common underlying cause and can present as a true emergency because of the risk of respiratory
Figure 1. CT scan demonstrating SVC obstruction (arrow).
distress [14]. However, in most adults there is time for additional investigation to confirm the diagnosis and establish the cause. Noninvasive methods such as Doppler ultrasound, CT scanning, MRI, phlebescintigraphy, and venography can be used to determine the size and location of obstructing lesions. Sputum cytology, thoracentesis, bronchial washings, bronchial biopsy, mediastinoscopy, and thoracotomy have been used to obtain tissue for pathologic diagnosis. In the present series, CT scan of the chest and superior venacavography were the two most frequently used imaging modalities for localizing the site occlusion. Both had diagnostic accuracies of 100%. Thoracotomy and bronchoscopy were usually effective in obtaining tissue for diagnosis, although we did find bronchoscopy to be unhelpful a third of the time. This is consistent with results previously reported. Shimm and co-workers [15] found diagnostic yields of 33% for sputum cytology, 33% for thoracentesis, 46% for bronchoscopy, and 100% for mediastinoscopy and thoracotomy.
Figure 2. Chest roentgenogram and superior venacavogram of SVC obstruction.
THE AMERICAN JOURNAL OF SURGERY
VOLUME 160 AUGUST 1990
209
CHENET AL
Figure 3. MRI scan showing a mass lesion compressing the SVC
(arrow).
SVC Syndrome
CT scan of Chest
/
\
No Mediastinal Mass
Mediastinal Mass with SVCO
Re-evaluation of Diagnosis
Sputum Cytology, Thoracentesis Needle Aspirate, Bronchoscopy
/ Definitive Malignancy
Radiatlon/Chemotherapy
\ Inconclusive Diagnosis
Superior Venacavography
Mediastinoscopy, Thoracotomy
Malignant Disease
Benign Disease
1
Bypass Surgery
Figure 4. Algorithm for evaluation of SVC syndrome.
Recent reviews [3,4,7,16] have shown that SVC syndrome is due to malignancy in over 85% of the cases. This led Lokich and Goodman [17] to propose therapeutic intervention prior to tissue biopsy. In their reported experience, the delay of diagnostic procedures led to nearfatal complications such as respiratory insufficiency, aspiration, and hemorrhage. In contrast, Davis and Shevland [11] and Shimm et al [15] found that invasive procedures can be carded out with minimal complications. Davis and Shevland [11] performed venography in 210
76 patients with SVC syndrome with 1 report of a minor complication of an inflammatory reaction at the puncture site. Shimm et al [15], in a series of 18 patients, showed that invasive diagnostic procedures can be carried out safely. In the present study, 33 patients with SVC syndrome were fully evaluated without any evidence of acute decompensation. The information obtained from the histologic diagnosis altered treatment plans in seven patients. Appropriate chemotherapeutic agents were administered to two patients with lymphoma, one with breast carcinoma, and one with oat cell carcinoma. Of the three patients with benign diseases, one was treated with anticoagulation and the other two with bypass grafting. There is no definitive surgical technique for correction of SVC syndrome. Lack of uniform success has contributed to a negative attitude toward surgical therapy for SVC syndrome. Earlier reviews by Gomes and Hufnagel [18] pointed out that although various surgical techniques have been used, the ideal substitute has yet to be discovered. Scherck et al [19] noted that autogenous, homologous, or heterologous vein, prosthetic grafts, and heterologous aortas have all been used as conduits for SVC obstruction bypass with variable success. Since these reviews were published, Avasthi and Moghissi [20] reported palliation of SVC syndrome for more than 5 months using Dacron grafts in four patients with inoperable bronchogenic carcinoma. Gladstone et al [9] presented two cases of venous decompression using a segment of autogenous femoral vein graft. Doty [21] reported successful long-term treatment of SVC syndrome with autogenous spiral vein grafts. Through the use of a composite vein graft made from the greater saphenous vein, two patients in the present series have undergone successful bypass. Early regression of symptoms was achieved, and there is no known graft failure at 3 years. Malignant lung cancers account for most cases of SVC syndrome in modern times. This has led some authors to suggest empiric radiotherapy without a need for definitive tissue diagnosis [17]. We believe this is excessively cavalier. The mortality associated with SVC syndrome does not relate to cavaI obstruction, but rather to the underlying cause. As Mahajan and co-workers [22] have elucidated, SVC syndrome is rarely an emergency because patients with caval obstruction have survived up to 28 years without medical treatment. In 33 patients with SVC syndrome, diagnostic procedures were performed without any major complications at our institution. We were able to establish a cause by the use of thoracotomy, bronchoscopy, needle biopsy, and/or mediastinoscopy. The treatment of seven patients was modified because of the tissue diagnosis. Systemic chemotherapeutic agents and surgical correction replaced supervoltage radiation as the primary mode of therapy in these patients. We believe that SVC syndrome should be approached diagnostically, like any other lung lesion. An algorithm for evaluation and therapy is depicted in Figure 4. If the syndrome is suspected, a CT scan of the chest is obtained for confirmation. Upon identification of a mediastinal obstructing lesion, sputum cells and/or pleural fluid is
THE AMERICANJOURNALOF SURGERY VOLUME160 AUGUST1990
SUPERIOR VENACAVASYNDROME
sent for cytologic examination. If tissue diagnosis remains inconclusive, bronchoscopy and/or fine needle aspiration is performed. If doubt still exists, then a superior venacavogram is obtained prior to tissue biopsy via mediastinoscopy or thoracotomy. At surgery, if no malignancy is identified, a bypass is performed with autogenous vein. We believe that all patients with SVC syndrome deserve a complete evaluation in order to optimize patient care in an era of expanding chemotherapeutic agents and improving vascular techniques. REFERENCES 1. Spencer FC. Disease of great vessels. In: Schwartz SI, Shires GT, Spencer FC, eds. Principles of surgery. 5th ed. San Francisco: McGraw-Hill, 1989. 2. Hunter W. The history of an aneursym of the aorta with some remarks on aneurysms in general. Med Observ Inq 1757; 1: 323-57. 3. Mclntire FT, Sykes EM. Obstruction of the superior vena cava: a review of the literature and report of two personal eases. Ann Intern Med 1949; 30: 925-60. 4. Fincher RE. Superior vena cava syndrome: experience in a teaching hospital. South Med J 1987; 80: 1243-5. 5. Parish JM, Marschke RF, Dines DE, Lee RE. Etiologic consideration in superior vena cava syndrome. Mayo Clin Proc 1981; 36: 407-13. 6. Kamiya K, Nakata Y, Naiki K, Hayashi H. Superior vena cava syndrome. J Vase Dis 1967; 4: 59-65. 7. Schraufnagel DE, Hill R, Leech JA, Pare JAP. Superior vena caval obstruction. Is it a medical emergency? Am J Med 1981; 70: 1169-74. 8. Mitchell IM, Saunders NR, Maber O, Lennox SC, Walker DR. Surgical treatment of idiopathic mediastinal fibrosis: report of five eases. Thorax 1986; 41: 210-4. 9. Gladstone DJ, Pillai R, Paneth M, Lincoln JCR. Relief of superior vena caval syndrome with autologous femoral vein used as a bypass graft. J Thorae Cardiovasr Surg 1985; 89: 750-2. 10. Reddy PJ, Cheung PT, Powell RC, Tierney WM. Obstruction of the superior vena r and trachea from benign goiters. Indiana Med 1986; 79: 1047-9. 11. Davis PF, Shevland JE. Superior vena caval obstruction: an analysis of seventy-sixcases, with comments on the safety of venography. Angiology 1985; 36: 354-7. 12. Roels P, Vincken W, DeGreve J, Vanhaelst L. Superior vena
cava syndrome caused by benign intrathoracic goiter. Acta Clin Belg 1983; 38: 329-32. 13. Light AM. Idiopathic fibrosis of mediastinum: a discussion of three cases and review of the literature. J Clin Pathol 1978; 31: 7888. 14. D'angio GJ, Mitos A, Evans AE. The superior mediastinal syndrome in children with cancer. AJR 1965; 93: 537-44. 15. Shimm DS, Logue GL, Rigsby LC. Evaluating the superior vena cava syndrome. JAMA 1981; 245: 951-3. 16. Lochridge SK, Knibbe WP, Doty DB. Obstruction of the superior vena cava. Surgery 1979; 85: 14-24. 17. Lokieh J J, Goodman R. Superior vena cava syndrome. Clinical management. JAMA 1975; 231: 58-61. 18. Gomes MN, Hufnagel CA. Superior vena cava obstruction. Ann Thorac Surg 1975; 20: 344-59. 19. Scherck JP, Kerstein MD, Stansel HC. The current status of vena caval replacement. Surgery 1974; 76: 209-33. 20. Avasthi RB, Moghissi K. Malignantobstruction of the superior vena cava and its palliation. J Thorac Cardiovasc Surg 1977; 74: 244-8. 21. Doty DB. Bypass of superior vena cava. J Thorac Cardiovasc Surg 1982; 83: 326-38. 2 2 . Mahajan V, Strimlan V, VanOrdstrand HS. Benign superior vena cava syndrome. Chest 1975; 68: 32-5.
DISCUSSION Thomas O'Donnell (Boston, MA): What is the role of anticoagulation in the management of this syndrome? Do you think you might extend the ability to reconstruct if you used polytetrafluoroethylene (PTFE) instead of spiral veins? John C. Chen (closing): Anticoagulation may be appropriate, but most of these patients have long-standing obstruction. Heparin does not appear to have a role in reversing these obstructions, although streptokinase and urokinase may. With regard to PTFE, although several authors have reported it to be less thrombogenic used in conjunction with aspirin and dipyridamole therapy, its role in this syndrome has yet to be evaluated.
THE AMERICAN JOURNAL OF SURGERY
VOLUME 160 AUGUST 1990 211
uperior vena cava (SVC) obstruction, although rare, can be responsible for life-threatening laryngeal edema and cerebral congestion. Serious sequelae of respiratory distress, seizures, and intracranial venous thrombosis have been associated with atypical presentations of SVC syndrome [1]. In the majority of cases, however, slow progressive occlusion of the SVC produces mild symptoms of headache, blurry vision, and neck and facial swelling. Collateral circulation generally enlarges ade-
S
From the Department of Surgery, Harbor-UCLA Medical Center, UCLA Schoolof Medicine,Torrance,California. Requestsfor reprintsshouldbe addressedto StanleyR. Klein,MD, Department of Surgery, Harbor-UCLA Medical Center, UCLA Schoolof Medicine,1000WestCarsonStreet, Box 15, Torrance,California 90509. Presented at the 18th AnnualMeetingof the Societyfor Clinical VascularSurgery,Palm Desert,California,March7-11, 1990.
quately to alleviate any transient disabilities due to acute vena caval insufficiency. Since Hunter's [2] first authentic description of SVC syndrome due to a luetic saccular aortic aneurysm, the spectrum of causes and diagnostic modalities has evolved drastically. Prior to 1949, only one third of the SVC occlusions reported were caused by primary chest malignancies. Mclntire and Sykes [3] reported their series of 502 cases in which 67% of patients with SVC syndrome had a benign cause and only 33% had underlying malignancies. This contrasts with Fincher's [4] 1987 experience in which 87% of patients with SVC syndrome had a primary intrathoracic cancer. In this retrospective study, we attempt to identify the frequent presenting findings and common causes of SVC syndrome and the useful diagnostic modalities and therapeutic outcome of various treatment regimens at our institution over the past decade, MATERIAL AND METHODS Between November 1976 and December 1988, 29 men and 16 women were treated for SVC syndrome at the Harbor-UCLA Medical Center. Their ages ranged from 18 to 76 years with a mean age of 54 years. This includes those admitted for a different diagnosis who were found subsequently to have stigmata of SVC syndrome. The initial symptoms consisted of dyspnea (71%), headache (11%), dysphagia (9%), cough (4%), and visual disturbance (4%). Physical findings at time of presentation are listed in Table I. More than half of the patients had facial engorgement, upper extremity edema, and shortness of breath on exertion. Less common signs were jugular venous distension, plethora, upper torso swelling, and unclear vision. The underlying causes of SVC syndrome in this series are listed in Table II. SVC syndrome was caused by malignant diseases in 42 of 45 patients (93%). Primary lung cancers were the leading cause, with squamous cell carcinoma accounting for 24% of the total. Metastatic breast carcinoma (11%) was the second most common cause, while lymphoma (9%) and seminoma (6%) were less common. Unusual malignant lesions that produced SVC syndrome were metastatic colon and renal cell carcinomas. Three patients had benign diseases (7%). One of these was a man who presented with a 2-year history of facial swelling, plethora, and upper-extremity edema. After an extensive work-up, he was found to have a fibrous mediastinal mass at sternotomy. Pathologic examination showed evidence of mediastinal fibrosis without signs of malignancy. The patient was given a final diagnosis of idiopathic mediastinal fibrosis. Another patient was a woman with a history of tuberculosis who presented with 4 weeks of cyanosis and facial edema. After an inconclu-
THE AMERICAN JOURNAL OF SURGERY VOLUME160 AUGUST 1990 207
CHEN ET AL
TABLE I Physical Findings at Presentation In 45 Patients Patients Findings
n
%
Facial/neck edema Extremity edema Dyspnea on exertion Jugular venous distension Plethora Chest/shoulder swelling Blurry vision
35 34 32 12 6 3 2
78 75 71 27 13 7 4
TABLE l I Cause of SVC Obstruction In 45 Patients
Patients Cause
n
%
Lung carcinoma Squamous cell carcinoma Undifferentiated carcinoma Adenocarclnoma Oat cell carcinoma Breast carcinoma Lymphoma Seminoma Malignant thymoma Metastatic colon carcinoma Metastatic renal cell carcinoma Benign idiopathic medlastlnal fibrosis Tuberculosis lymphangitis Epicardial pacemaker wire
11 7 5 3 5 4 3 2 1 1 1 1 1
24 16 11 7 11 9 6 4 2 2 2 2 2
TABLE I I I
Diagnostic Modalltles Used In the Evaluation of SVC Syndrome Procedure
No. Done
Chest computerized tomography Superior venacavogram Thoracotomy/sternotomy Bronchoscopy Barium swallow Needle aspiration Radionuclide scan* Magnetic resonance imaging Medlastinoscopy Upper endoscopy Clinical diagnosis
23 12 6 6 4 2 2 2 1 1 12
* Technetium-99m.
sivr search, she was presumed to have tuberculosis lymphadenitis. The third patient complained of dyspnea on exertion for 6 weeks and right-sided facial edema lasting 8 months. He underwent a superior venacavogram and was found to have a small area of obstruction (narrowing) that ultimately proved to be an improperly placed epicardial pacemaker wire. Fifty-nine diagnostic procedures were performed on 33 patients to define the cause and location of their obstruction. Eleven of 45 patients underwent treatment 208
THE AMERICAN JOURNAL OF SURGERY
based on history and clinical findings alone. The procedures used are listed in Table III. The diagnostic modality most frequently employed was chest computed tomography (CT). CT scanning confirmed SVC occlusion in all 23 patients in whom it was performed. The criteria for a positive CT scan included poor opaciflcation of the SVC with intravenous contrast injection, evidence of a mediastinal mass extending into the SVC, and numerous venous collaterals demonstrated on the lateral and anterior chest wall (Figure 1). Venacavography was used in 12 patients. Superior venacavogram accurately defined the site of obstruction in all those studied (Figure 2). There were no reported complications from this procedure. Thoracotomy or sternotomy was necessary in 6 of the 45 patients (13%) with SVC syndrome, Surgery was performed solely for the purpose of obtaining tissue for diagnosis in four of the patients. The remaining two patients underwent SVC bypass grafting and excisional biopsy at the same operation. Bronchoscopy was useful in diagnosis 67% of the time; four of the six bronchoscopies performed yielded positive cytology for malignancy. Radionuclide scanning and magnetic resonance imaging (MRI) were infrequently used, but both were effective in confirming SVC obstruction. MRI showed a mediastinal mass compressing the SVC in the single patient in whom it was performed (Figure 3). Of the 42 patients with malignant disease, 33 had relief of symptoms with radiation and/or chemotherapy, 7 were lost to follow-up, and 2 died prior to initiation of therapy. Of the three patients with benign diseases, two underwent SVC reconstruction with immediate resolution of symptoms. Autologous saphenous vein grafts were used as bypass conduits in both patients. These two patients had no evidence of SVC syndrome at 3-year followup. In the third patient who sustained SVC thrombosis due to a cardiac pacing wire, anticoagulation was sufficient in providing long-term symptomatic relief. However, he died 2 years later of progressive cardiomyopathy. COMMENTS SVC syndrome is a rare clinical entity first described by William Hunter in 1757 [2]. It has a distinctive presentation that can be easily recognized once the diagnosis is entertained. Parish and co-workers [5] maintain that physical examination is all that is required in most patients. Occlusion of the venous outflow from the head, neck, and upper extremities produces characteristic findings of facial edema, cyanosis, dyspnea, and prominent neck veins that hallmark the syndrome. Obstruction can be due to thrombosis, extrinsic mass effect, or invasion of venous intima by neoplasm. Recently, an important change has occurred in the common causes of SVC syndrome. A 1949 review by McIntire and Sykes [3] showed that of 502 cases of SVC syndrome analyzed, 38% were caused by primary thoracic tumors, 30% by aortic aneurysm, 15% by chronic mediastinitis, 8% by phlebitis, 6% by metastatic disease, and 3% by an uncertain cause. Kamiya and colleagues [6] studying 734 patients from 1949 to 1965 demonstrated that 540 (74%) of them had intrathoracic malignancy as the cause of SVC syndrome.
VOLUME 160
AUGUST 1990
SUPERIOR VENA CAVASYNDROME
In recent surveys by Fincher [4] and Schraufnagel et al [7], malignant diseases accounted for 87% and 97% of the cases, respectively. Our series found a malignant cause in 93% of the cases. This trend toward malignancy is most likely a result of improved detection and treatment of granulomatous and infectious mediastinal lesions before they progress to obstruction. Benign causes of SVC syndrome have been relatively uncommon. With the declining cases of tuberculosis and syphilis, curable SVC obstructions are infrequent. Mitchell and co-workers [8] reported their experience using autogenous saphenous vein grafts to perform successful bypasses in two patients with benign idiopathic mediastinal fibrosis. Gladstone et al [9] described effective bypass in one patient with mediastinal fibrosis using the femoral vein as a conduit. Reddy et al [10], Davis and Shevland [11], and Roels et al [12] have each reported two cases of SVC syndrome due to benign goiters. Three new cases of benign eaval obstruction are discussed in the present study. One patient developed obstruction of the SVC due to idiopathic mediastinal fibrosis, a rare fibrosclerotic condition of unknown cause. Various factors have been implicated in its pathogenesis, in particular infections with tuberculosis and histoplasmosis, the drug methysergide, and autoimmune processes [13]. To date, however, absolute cause and effect has not been demonstrated with any single agent. Another patient was diagnosed as having tuberculosis lymphadenitis. Tuberculosis infection evokes a reactive cicatricial contraction in the mediastinum, with gradual obstruction of the SVC [3]. The third patient was found to have SVC syndrome following an epieardial pacemaker wire insertion. With the marked decrease in infectious lesions of the mediastinum and the recent increased use of invasive monitors such as SwanGanz catheters, pacemaker electrodes, and central alimentation catheters, iatrogenic causes may emerge as a common cause of nonmalignant SVC syndrome in the future. SVC syndrome often causes tracheal compression and airway compromise in the pediatric population. Lymphoma is the most common underlying cause and can present as a true emergency because of the risk of respiratory
Figure 1. CT scan demonstrating SVC obstruction (arrow).
distress [14]. However, in most adults there is time for additional investigation to confirm the diagnosis and establish the cause. Noninvasive methods such as Doppler ultrasound, CT scanning, MRI, phlebescintigraphy, and venography can be used to determine the size and location of obstructing lesions. Sputum cytology, thoracentesis, bronchial washings, bronchial biopsy, mediastinoscopy, and thoracotomy have been used to obtain tissue for pathologic diagnosis. In the present series, CT scan of the chest and superior venacavography were the two most frequently used imaging modalities for localizing the site occlusion. Both had diagnostic accuracies of 100%. Thoracotomy and bronchoscopy were usually effective in obtaining tissue for diagnosis, although we did find bronchoscopy to be unhelpful a third of the time. This is consistent with results previously reported. Shimm and co-workers [15] found diagnostic yields of 33% for sputum cytology, 33% for thoracentesis, 46% for bronchoscopy, and 100% for mediastinoscopy and thoracotomy.
Figure 2. Chest roentgenogram and superior venacavogram of SVC obstruction.
THE AMERICAN JOURNAL OF SURGERY
VOLUME 160 AUGUST 1990
209
CHENET AL
Figure 3. MRI scan showing a mass lesion compressing the SVC
(arrow).
SVC Syndrome
CT scan of Chest
/
\
No Mediastinal Mass
Mediastinal Mass with SVCO
Re-evaluation of Diagnosis
Sputum Cytology, Thoracentesis Needle Aspirate, Bronchoscopy
/ Definitive Malignancy
Radiatlon/Chemotherapy
\ Inconclusive Diagnosis
Superior Venacavography
Mediastinoscopy, Thoracotomy
Malignant Disease
Benign Disease
1
Bypass Surgery
Figure 4. Algorithm for evaluation of SVC syndrome.
Recent reviews [3,4,7,16] have shown that SVC syndrome is due to malignancy in over 85% of the cases. This led Lokich and Goodman [17] to propose therapeutic intervention prior to tissue biopsy. In their reported experience, the delay of diagnostic procedures led to nearfatal complications such as respiratory insufficiency, aspiration, and hemorrhage. In contrast, Davis and Shevland [11] and Shimm et al [15] found that invasive procedures can be carded out with minimal complications. Davis and Shevland [11] performed venography in 210
76 patients with SVC syndrome with 1 report of a minor complication of an inflammatory reaction at the puncture site. Shimm et al [15], in a series of 18 patients, showed that invasive diagnostic procedures can be carried out safely. In the present study, 33 patients with SVC syndrome were fully evaluated without any evidence of acute decompensation. The information obtained from the histologic diagnosis altered treatment plans in seven patients. Appropriate chemotherapeutic agents were administered to two patients with lymphoma, one with breast carcinoma, and one with oat cell carcinoma. Of the three patients with benign diseases, one was treated with anticoagulation and the other two with bypass grafting. There is no definitive surgical technique for correction of SVC syndrome. Lack of uniform success has contributed to a negative attitude toward surgical therapy for SVC syndrome. Earlier reviews by Gomes and Hufnagel [18] pointed out that although various surgical techniques have been used, the ideal substitute has yet to be discovered. Scherck et al [19] noted that autogenous, homologous, or heterologous vein, prosthetic grafts, and heterologous aortas have all been used as conduits for SVC obstruction bypass with variable success. Since these reviews were published, Avasthi and Moghissi [20] reported palliation of SVC syndrome for more than 5 months using Dacron grafts in four patients with inoperable bronchogenic carcinoma. Gladstone et al [9] presented two cases of venous decompression using a segment of autogenous femoral vein graft. Doty [21] reported successful long-term treatment of SVC syndrome with autogenous spiral vein grafts. Through the use of a composite vein graft made from the greater saphenous vein, two patients in the present series have undergone successful bypass. Early regression of symptoms was achieved, and there is no known graft failure at 3 years. Malignant lung cancers account for most cases of SVC syndrome in modern times. This has led some authors to suggest empiric radiotherapy without a need for definitive tissue diagnosis [17]. We believe this is excessively cavalier. The mortality associated with SVC syndrome does not relate to cavaI obstruction, but rather to the underlying cause. As Mahajan and co-workers [22] have elucidated, SVC syndrome is rarely an emergency because patients with caval obstruction have survived up to 28 years without medical treatment. In 33 patients with SVC syndrome, diagnostic procedures were performed without any major complications at our institution. We were able to establish a cause by the use of thoracotomy, bronchoscopy, needle biopsy, and/or mediastinoscopy. The treatment of seven patients was modified because of the tissue diagnosis. Systemic chemotherapeutic agents and surgical correction replaced supervoltage radiation as the primary mode of therapy in these patients. We believe that SVC syndrome should be approached diagnostically, like any other lung lesion. An algorithm for evaluation and therapy is depicted in Figure 4. If the syndrome is suspected, a CT scan of the chest is obtained for confirmation. Upon identification of a mediastinal obstructing lesion, sputum cells and/or pleural fluid is
THE AMERICANJOURNALOF SURGERY VOLUME160 AUGUST1990
SUPERIOR VENACAVASYNDROME
sent for cytologic examination. If tissue diagnosis remains inconclusive, bronchoscopy and/or fine needle aspiration is performed. If doubt still exists, then a superior venacavogram is obtained prior to tissue biopsy via mediastinoscopy or thoracotomy. At surgery, if no malignancy is identified, a bypass is performed with autogenous vein. We believe that all patients with SVC syndrome deserve a complete evaluation in order to optimize patient care in an era of expanding chemotherapeutic agents and improving vascular techniques. REFERENCES 1. Spencer FC. Disease of great vessels. In: Schwartz SI, Shires GT, Spencer FC, eds. Principles of surgery. 5th ed. San Francisco: McGraw-Hill, 1989. 2. Hunter W. The history of an aneursym of the aorta with some remarks on aneurysms in general. Med Observ Inq 1757; 1: 323-57. 3. Mclntire FT, Sykes EM. Obstruction of the superior vena cava: a review of the literature and report of two personal eases. Ann Intern Med 1949; 30: 925-60. 4. Fincher RE. Superior vena cava syndrome: experience in a teaching hospital. South Med J 1987; 80: 1243-5. 5. Parish JM, Marschke RF, Dines DE, Lee RE. Etiologic consideration in superior vena cava syndrome. Mayo Clin Proc 1981; 36: 407-13. 6. Kamiya K, Nakata Y, Naiki K, Hayashi H. Superior vena cava syndrome. J Vase Dis 1967; 4: 59-65. 7. Schraufnagel DE, Hill R, Leech JA, Pare JAP. Superior vena caval obstruction. Is it a medical emergency? Am J Med 1981; 70: 1169-74. 8. Mitchell IM, Saunders NR, Maber O, Lennox SC, Walker DR. Surgical treatment of idiopathic mediastinal fibrosis: report of five eases. Thorax 1986; 41: 210-4. 9. Gladstone DJ, Pillai R, Paneth M, Lincoln JCR. Relief of superior vena caval syndrome with autologous femoral vein used as a bypass graft. J Thorae Cardiovasr Surg 1985; 89: 750-2. 10. Reddy PJ, Cheung PT, Powell RC, Tierney WM. Obstruction of the superior vena r and trachea from benign goiters. Indiana Med 1986; 79: 1047-9. 11. Davis PF, Shevland JE. Superior vena caval obstruction: an analysis of seventy-sixcases, with comments on the safety of venography. Angiology 1985; 36: 354-7. 12. Roels P, Vincken W, DeGreve J, Vanhaelst L. Superior vena
cava syndrome caused by benign intrathoracic goiter. Acta Clin Belg 1983; 38: 329-32. 13. Light AM. Idiopathic fibrosis of mediastinum: a discussion of three cases and review of the literature. J Clin Pathol 1978; 31: 7888. 14. D'angio GJ, Mitos A, Evans AE. The superior mediastinal syndrome in children with cancer. AJR 1965; 93: 537-44. 15. Shimm DS, Logue GL, Rigsby LC. Evaluating the superior vena cava syndrome. JAMA 1981; 245: 951-3. 16. Lochridge SK, Knibbe WP, Doty DB. Obstruction of the superior vena cava. Surgery 1979; 85: 14-24. 17. Lokieh J J, Goodman R. Superior vena cava syndrome. Clinical management. JAMA 1975; 231: 58-61. 18. Gomes MN, Hufnagel CA. Superior vena cava obstruction. Ann Thorac Surg 1975; 20: 344-59. 19. Scherck JP, Kerstein MD, Stansel HC. The current status of vena caval replacement. Surgery 1974; 76: 209-33. 20. Avasthi RB, Moghissi K. Malignantobstruction of the superior vena cava and its palliation. J Thorac Cardiovasc Surg 1977; 74: 244-8. 21. Doty DB. Bypass of superior vena cava. J Thorac Cardiovasc Surg 1982; 83: 326-38. 2 2 . Mahajan V, Strimlan V, VanOrdstrand HS. Benign superior vena cava syndrome. Chest 1975; 68: 32-5.
DISCUSSION Thomas O'Donnell (Boston, MA): What is the role of anticoagulation in the management of this syndrome? Do you think you might extend the ability to reconstruct if you used polytetrafluoroethylene (PTFE) instead of spiral veins? John C. Chen (closing): Anticoagulation may be appropriate, but most of these patients have long-standing obstruction. Heparin does not appear to have a role in reversing these obstructions, although streptokinase and urokinase may. With regard to PTFE, although several authors have reported it to be less thrombogenic used in conjunction with aspirin and dipyridamole therapy, its role in this syndrome has yet to be evaluated.
THE AMERICAN JOURNAL OF SURGERY
VOLUME 160 AUGUST 1990 211
