Superior Vena Cava Syndrome The Myth - the Facts1,2
ALON YELLIN, ADA ROSEN, NIRA REICHERT, and YAIR LIEBERMAN
Introduction Superior vena cava syndrome (SVCS) is the clinical presentation of SVC obstruction, or severe reduction in venous return from the head, neck, and upper extremities. Many mediastinal conditions may cause obstruction of the SVC either by compression or invasion, and some by thrombus formation (1). Yet for many, SVCS is almost synonymous with bronchogenic carcinoma, usually of the small cell anaplastic type. SVCS is regarded as an oncologic emergency, even in recent textbooks (2), and therapy is often initiated before a diagnosis is established. Invasive diagnostic procedures are considered hazardous by many (2-5), mainly because of excessive bleeding and respiratory complications. It was our impression that many pa-tients seen by us had conditions other than bronchogenic carcinoma. We could not recall a patient in acute distress caused by the SVCS per se. We also had to surgically obtain a tissue diagnosis in many patients in whom the SVCS was the presenting symptom of their mediastinal disease, and could not recall any outstanding complications. Bearing these controversies in mind, we reviewed our experience with SVCS during a 16-yr period. Methods During the 16 yr between 1972 and 1987, a total of 63 patients with SVCS were treated at the Sheba Medical Center, Israel. These patients were located by a computer search of the archives, and a manual search of the surgical list. All charts were retrieved, and pertinent data were recorded. Patients with venous congestion caused by cardiac failure or pericardial diseases without genuine SVC obstruction were excluded. Patients were followed until death or until early 1988 except for eight patients who could not be located. Statistical analysis consisted of a contingency table analysis using the Apple Stats Plus software. The series included 63 subjects. There was a steady increase in the annual incidence of 1114
SUMMARY This retrospective review was performed with an intent to clarify several misconceptions associated with superior vena cava syndrome (SVCS). During a 16-yr period, we diagnosed and treated 63 patients with SVCS,including seven (11%)children younger than 10 yr of age. Thirty patients (47.6%) had bronchogenic carcinoma, and 13 (20.6%) had lymphoma. In 43 cases SVCS was the presenting symptom of a mediastinal condition. Forty-one patients underwent diagnostic procedures with no major complications, and diagnosis was obtained In 36. Only six patients had surgical treatment, and 45 had radiation therapy, chemotherapy, or both. There was no mortality associated directly with venous congestion. Symptomatic relief occurred in 80% of treated patients. Weconclude that: (1)SVCSper se should not be feared (symptomatic relief is the rule), (2) accurate diagnosis can be achieved with minimal morbidity, and (3) the versatile underlying etiology dictates the outcome that can be Improved with appropriate therapy. AM REV RESPIR DIS 1990; 141:1114-1118
SVCS, though during the entire period the annual load of hospitalization had not changed. For the first period of the study (1972 to 1977)the average incidence was 1.5cases/yr. From 1978to 1983,it increased to 4.0 cases/yr, and from 1984 to 1987, to 7.5 cases/yr. There were 49 men and 14 women ranging in age from 1 to 80 yr (mean, 45 yr) (figure 1), including seven children younger than 10 yr of age. The majority were in the sixth and seventh decades of life.
Clinical Presentation The most common presenting symptoms were edema and venous congestion of the head and neck or the upper extremities followed by various respiratory complaints (table 1).Only five patients presented with hemoptysis. Chest pain, dysphagia, and speech difficulties were uncommon. None of our patients presented with, or developed, neurologic symptoms except when brain metastases were present. In 43 of 63 patients (68070), the SVCS was the presenting symptom of their disease. Their symptomatology was similar to that of patients whose SVCS developed in the course of a known disease. The duration of symptoms before admission was relatively long (table 2). In 43% of the patients, it was 5 days to 1 month, but in the remaining 34 it ranged from 1 month to 4 yr (mean, 5.5 months). Patients with malignant disease had a shorter duration of symptoms (average, 45 days) than did patients with benign conditions (average, 420 days). From the latter group, the three patients with mediastinal tumors had the longest duration
(average, 3.5 yr), whereas patients with SVC thrombosis had the shortest duration (mean, 15 days).
Underlying Conditions and Diagnostic Procedures Twenty of the patients were known to have a disease that eventually led to SVC obstruction (table 3). Of these, half had bronchogenic carcinoma, one had a lymphoma, two had primary malignant mediastinal tumors, and in three, metastatic spread from breast cancer caused the SVCS. In only four of these 20 patients was the underlying cause benign, including three children after the Mustard procedure, and one adult with SVC thrombosis after implantation of intravenous pacemaker electrodes. Of the remaining 43 patients, 41 were admitted for evaluation of SVCS, and in two the SVCS developed during hospitalization for an undetermined mediastinal condition. Bronchogenic carcinoma accounted for less than half of the cases. Lymphoma was responsible for one-fifth of all cases and for more than one-quarter of the new cases,
(Received in original form May 17, 1989 and in revised form October 17, 1989) 1 From the Department of Thoracic Surgery and Pulmonary Diseases, The Chaim Sheba Medical Center, Tel Hashomer, Israel. 2 Correspondence and requests for reprints should be addressed to Alon Yellin, M.D., Department of Thoracic Surgery, The Chaim Sheba Medical Center, Tel Hashomer 52621, Israel.
1115
THE MYTH AND FACTS OF SUPERIOR VENA CAVA SYNDROME
PTS 20
0
IS
Wales
_
Females
16 . 14 .
Fig. 1. Age distribution of patients with superior vena cava syndrome. The majority were 60 yr of age or older, corresponding to the age groups in which bronchogenic carcinoma is common.
12 10 .
0~.
JjJLD 0-10
11-20 21-30 31-40 41-50 51-60 61-70
non- Hodgkin's lymphoma underlying all but one case. Of the metastatic cases, those related to breast cancer were clinically suspected and radiologically compatible. In patients with a known malignant disease, histologic confirmation of the mediastinal mass was not sought prior to initiation of therapy. The finding of a metastatic bladder carcinoma on mediastinoscopy in one case was unexpected, as the patient was known to have a noninvasive bladder tumor. Nonmalignant conditions caused SVCS in II patients, including three cases of benign mediastinal tumors. In patients younger than 20, the condition underlying SVCS were either iatrogenic, i.e., after Mustard operation (three cases) or primary mediastinal neoplasms: lymphoma (five) and seminoma (one case). Among Arab patients, the rate of lymphoma (table 4) was significantly higher than among Jewish patients.
Diagnostic Procedures Twenty patients who had a known condition leading to SVCS were evaluated by means of TABLE 1 PRESENTING SYMPTOMS IN 63 PATIENTS WITH SVC SYNDROME Patients Symptoms
(n)
(%)
Venous congestion head and neck Swelling upper extremities or trunk Dyspnea, respiratory distress Cough, hoarseness Hemoptysis Chest pain Dysphagia Dysphonia
54 20 19 13 5 4 4
85.7 31.7 30.2 20.6 7.9 6.3 6.3 1.6
1
TABLE 2 DURATION OF SYMPTOMS BEFORE DIAGNOSIS OR THERAPY' Duration (months)
.. 1 1-6 7-12 > 12 Total
Unknown Cases
Known Diagnosis
15 25
11 5 1
3 43
17
Total
26 30 1 3 60
, n = 60; in three patients the exact duration of symptoms was unclear.
[j
>70 AGE (YRS)
conventional radiology and CT scans for anatomic localization. In the malignant cases- 16 altogether - the histologic nature of the mediastinal involvement was assumed identical to the original malignancy, and surgical procedures to prove it were usually not undertaken. In new cases, diagnosis was considered crucial (table 5). All patients had chest radiographs taken, and either linear tomography or CT scans. All but two underwent further diagnostic procedures. One elderly patient with a long-standing history was assumed to have retrosternaI goiter, and another died prior to further investigation. Semi-invasiveprocedures were helpful in 52070 of the patients,
providing histologic (seven cases), cytological (three cases), or anatomic (two cases) diagnoses. Invasive procedures ranged from lymph node biopsy (five cervical, two axillary) through mediastinoscopy to median sternotomy and formal thoracotomy. These 27 procedures were associated with no mortality and no major bleeding, obtaining tissue diagnosis in 89% of all cases. Only fivepatients, including the one with retrosternal goiter, remained without a histologic diagnosis, which proved eventually to be bronchogenic carcinoma in four.
Results
Therapy The aim of management was to alleviate symptoms and prolong life when possible in the noncurable malignant cases, treat aggressively the favorable malignant neoplasm, and achieve cure in the benign conditions. Therefore, surgery was reserved for patients with treatable vascular problems (after Mustard operation or SVC thrombosis), or benign mediastinal tumors (one goiter, one fibrous mesothelioma). Complete resection was not attempted in cases of malignant tumors in-
TABLE 3 CONDITIONS UNDERLYING SUPERIOR VENA CAVA SYNDROME (SVCS)' Newly Diagnosed Cases Tumors
(n)
(%)
Cases with Known Diagnosis (n)
20
46.5
10
Malignant Bronchogenic carcinoma Adenocarcinoma Squamous cell Small cell Large cell Anaplastic Undetermined Lymphoma Hodgkin's disease Non-Hodgkin's lymphoma Primary mediastinal tumors Thymoma Seminoma Esophageal carcinoma Metastatic tumors Breast Bladder Undetermined Total malignant
1t 1 36
Benign SIP mustard operation SVC thrombosis Retrosternal goiter Mediastinal fibrosis Behcel's syndrome Fibrous mesothelioma Total benign Grand total
2 2 1 1 1 7 43
All Cases (n)
(%)
47.6
4
2
30 6
3 4 2 3 4 12
3
6
1 11 2 1 1
4
2 1
4 27.9
1 1
4.7
2 1 1
3 3
4 8 13 2 11 4 2 1 1 4
16
52
3
3 3 2
1
16.3
4 20
6.3
6.3
3 1 1
83.7
20.6
1 1 1 11
1.6 82.5
17.5
63
, The incidence of lymphoma in previously undiagnosed cases is higher than in the entire series, and certainly higher than in patients with a known condition. t The diagnosis of bladder carcinoma was known, but the SVCS was believed unrelated.
1116
YELLIN, ROSEN, REICHERT, AND LIEBERMAN
Eighty percent of nonterminal patients achieved symptomatic improvement or complete resolution of the SVCS. In at least five, the improvement had begun before any therapy was initiated. Symptomatic relief was independent of the management. In treated patients, resolution of symptoms was unrelated to the nature of the disease. Eight of the 11 patients with benign disease (73070) achieved relief compared with 36 of the 52 (69%) patients with malignant disease (including terminal cases). With lung cancer (20 patients out of 30) the symptomatic improvement was less pronounced (11 out of 13)than in lymphoma (66% and 85%, respectively; p = NS).
TABLE 4 HISTOLOGY VERSUS ORIGIN IN SUPERIOR VENA CAVA SYNDROME"
Jew Arab Total
Cancer
Lymphoma
Other
Total
p Value
26
7 6 13
18
51 12 63
< 0.05
4
30
2
20
" Four of the six Arab patients with lymphoma were childrenyounger than 10 yr of age.
TABLE 5 PROCEDURES AIDING DIAGNOSIS IN NEW CASES OF SUPERIOR VENA CAVA SYNDROME" Condition Diagnosed Patients (n)
Diagnostic (n)
Semi-invasive Bronchoscopy Thoracentesis Venography Lymphangiography Dynamic scan Total semi-invasive
13 3 3 1 1 23
8t 2:1: 2§ 1 1§ 12
Invasive Lymph node Bx Suprasternal Bx Mediastinoscopy Mediastinotomy Median sternotomy Thoracotomy Total invasive Total procedures
7 1 15 1 2 1 27 50
511 1 14
•n
Lung Cancer
Lymphoma
6
2 2
Other
2
6
4
2
3
7
1 4
2
3 1 2
1
2 1 24 36
1
10 16
8
6
12
8
= 41.
t
Seven histology; one cytology. Both cytology. § Showed benign thrombosis secondary to mediastinal disease. I One fine needle aspiration.
*
vading the aorta or heart, nor did SVC resection or bypass seem appropriate. When the SVCS was clinically defined, all patients, except for five in terminal condition, were subjected to treatment (table 6). The type of management depended on the diagnosis. Lymphomas were treated with a curative intent, usually chemotherapy with or without radiotherapy. Patients with bronchogenic carcinomas, where the goal was to alleviate symptoms, were offered radiotherapy.
The combination of chemical agents or the amount of radiation were tailored according to the individual tissue diagnosis. Thrombolysis practiced in three patients consisted of streptokinase (two cases) and heparin infused continuously to an upper extremity vein. Medical modalities consisting of diuretics, corticosteroids, and systemic anticoagulation wereadministered to 31 patients, but were the sole treatment in two only, and therefore their efficacy cannot be evaluated.
Survival Thirty- four (54 %) patients have died (table 7), 18 during the first 6 months after the first clinical presentation, and 16patients from 6 months to 2 yr thereafter. Twenty-one (33 %) patients are alive, seven 1 yr or less after diagnosis and treatment, and 14one to seven years from the initial presentation. Patients with benign conditions had the best outcome, followed by patients with lymphoma. Twenty-five patients with bronchogenic carcinoma were followed, compared with 11 cases with lymphoma. The l-yr survival and median survival were 18%, 8.5 months, and 70%, not reached, respectively. Death in all patients was caused by the underlying disease and not by consequences of the SVCS. Discussion
Superior vena cava syndrome may result from many mediastinal conditions. It was first described by William Hunter in 1757 as a complication of a syphilitic aortic aneurysm (6). Although until the beginning of this century the most common etiologic entities associated with SVCS were benign, such as aortic aneurysms,
TABLE 6
TABLE 7
SYMPTOMATIC RESPONSE TO THERAPY
OUTCOME VERSUS ETIOLOGY
Relief Management Surgery Chemotherapy (CTx) Radiotherapy (RTx) CTx and RTx CTx, RTx, and lysis Thrombolysis Other medical None Total Terminal
Number
Followed
(n)
(%)
6 8 25 10 2 3 2 2 58 5
6 8 23 10 2 3 1 2 55
5 6 18 9 2 3 1 0
83 75 78 90
44
80
0
Alive
Etiology Benign Lung cancer Lymphoma Metastatic Mediastinal tumor Other malignancies Total
Months
Range (months)
8 5 7
51 13.2 25
13-120 7-24 4-65
21
35.3
Patients
• One was alive for at least5 months.
Dead Patients
Months
Range
3 20 4 3 3 1
7.7 10.2 8 26 5 2
0-12 0-96 2-18 13-52 1-8
34
10.4
Lost to Follow-up
5 2"
8
1117
THE MYTH AND FACTS OF SUPERIOR VENA CAVA SYNDROME
tuberculosis, and chronic mediastinitis, the pendulum has shifted towards the malignant lesions that constitute 85 to 90070 of all cases (I, 7-9). The most common malignant tumor causing SVC obstruction is bronchogenic carcinoma, which has been responsible for 67 to 82% of cases (1): neoplasms of the right lung (7) and small-cell lung cancer are more often implicated (1,9). According to Nieto and Doty (1), lymphomas underlie SVC in only 5 to 15% of the cases, and metastatic lesions in 3 to 20%. In children the most common causes are usually congenital heart disease and cardiovascular surgery, followed by lymphomas (7). In recent years there has been an increase in iatrogenic cases, i.e., those associated with insertion of pacemakers, SwanGanz catheters, and central venous lines (10). These rates are quite similar to those reported in the past from our institute by Adar and associates (11). In that study, which included patients from 1955 to 1972, there were only two patients with lymphoma. The annual incidence was 1.8 cases, which is similar to the incidence in the first period of our study. It is not entirely clear why the incidence increased progressively to 7.5 cases per year. It may be due to better awareness of this condition, more accurate recording, and a higher incidence of lung cancer. The highest proportion of patients with SVCS after the sixth decade certainly fits the age distribution of bronchogenic carcinoma. The present series includes a high proportion of patients with lymphoma in whom the SVCS heralded their disease. The modern therapy for lymphoma practically precludes the late occurrence of SVCS. Lymphoma accounted for 12% or less of the cases in other studies (8, 10, 12-15). Similar to the present experience and to our previous impression (16), Hodgkin's lymphoma was seldom a cause of SVCS. It can be argued that the high rate of lymphoma in our series may be partly due to selective referral of Arab children from the Gaza Strip. The incidence of T-Iymphoma, especially in the pediatric age group, is known to be high in that area. This probably explains why Arab patients had a much higher incidence oflymphoma than did Jewish patients, in whom the main etiology was bronchogenic carcinoma. On the other hand, it is noteworthy that SVCS is so rarely reported with malignant thymoma. In our practice thymoma (invasive and noninvasive) is more common than mediastinal lymphoma. The clinical syndrome is caused by
compression or obstruction of the SVC and altered venous return from the head, neck, and upper extremities. The severity of the SVC syndrome depends on the rapidity of occlusion and collateral vessel development - the more acute the occlusion, the more severe the syndrome (17). Collaterals drain to the azygos vein in cases of high obstruction, or to the hemiazygos or through chest wall veins to the inferior vena cava in cases of obstruction below the azygos confluence point (1). Yetin some cases the SVC may develop after surgical procedures as a consequence of destruction of mediastinal collateral vessels. Similarly, late development of collaterals may cause a spontaneous relief of symptoms without any treatment, as observed in five of our patients. In the past, and even recently, the complications of SVC obstruction, mainly laryngeal and cerebral edema, were feared (2,4). This led to the institution of emergency therapy, usually consisting of highdose radiation before tissue diagnosis was established (5, 14). Occasionally, heroic operations were performed to bypass the obstructed SVC and reduce the venous pressure (18), and even to resect what are usually considered nonoperable tumors (19). The attitude of unsound therapy also leaned on the presumed high morbidity associated with invasive procedures (2-5). Bronchoscopy should not be met with any outstanding complication unless the venous return from the lungs is also compromised. Bleeding occurring during lymph node biopsy or thoracotomy can be easily controlled. Mediastinoscopy seems hazardous, but as one is usually confronted with a limited number of dilated veins, the procedure can be accomplished by experienced surgeons with minimal complications. In our series there was only one case of minor bleeding, managed by compression. Ahmann (9) reviewed the literature and found a very low morbidity rate associated with diagnostic procedures. Recent studies (8, 12) have confirmed this observation. Nevertheless, it is justified to initiate therapy prior to precise histologic diagnosis in the occasional patient with life-threatening laryngeal or cerebral edema. Rarely, one is also confronted with a patient, usually a child, whose huge anterior mediastinal mass severely narrows the trachea (3, 20), posing a considerable anesthetic hazard. Diagnostic procedures can be delayed in such cases. It should be stressed again that the respiratory embarrassment is not due to the SVCS.
In concurrence with previous publications (7, 8, 12), we found no ground for the belief that SVCS per se frequently results in fatal complications. Moreover, not uncommonly, patients may improve spontaneously as a result of slowly enlarging collaterals. The underlying disease dictates the outcome and should therefore be managed appropriately. The span of conditions causing SVCS is wide, and management is not identical. Accurate diagnosis is crucial for planned therapy. With proper therapy a high rate of symptomatic relief can be expected, and even survival may approximate that of patients with similar diseases without SVCS (8). Long-term survivors have occasionally been reported, even in patients with bronchogenic carcinoma and SVCS (21, 22). We therefore recommend that patients presenting with superior vena cava syndrome should be carefully and promptly evaluated. Every effort should be made to obtain a precise tissue or anatomic diagnosis, and only then should specific therapy follow. Gratifying results can be expected. References 1. Nieto AF, Doty DB. Superior vena cava obstruction, clinical syndrome, etiology, and treatment. Curr Probl Surg 1986; 10:442-84. 2. Carabell SC, Goodman RL. Oncologic emergencies: superior vena cava syndrome. In: DeVita VT, Hellman S, Rosenberg SA, eds. Cancer. 2nd ed. Philadelphia: Lippincott, 1985; 1855-60. 3. Northrip DR, Bohman BK,Tsueda K. Total airway occlusion and superior vena cava syndrome in a child with an anterior mediastinal tumor. Anesth Analg 1986; 65:1079-82. 4. Salsali M, Cliff ton EE. Superior vena caval obstruction in carcinoma of lung. NY State J Med 1969; 69:2875-89. 5. Issa PY, Brihi ER, Janin Y, Slim MS. Superior vena cavasyndrome in childhood: report of ten cases and review of the literature. Pediatrics 1983; 71: 337-41. 6. Hunter W. The history of an aneurysm of the aorta with some remarks on aneurysm in general. Med Observ Inquir (Lond) 1957; 1:323. 7. Sculier JP, Feld R. Superior vena cava obstruction syndrome: recommendations for management. Cancer Treat Rev 1985; 12:209-18. 8. Schraufnagel DE, Hill R, Leech JA, Pare JAP. Superior vena caval obstruction: is it a medical emergency? Am J Med 1981; 70:1169-74. 9. Ahmann FR. A reassessment of the clinical implications of the superior vena cava syndrome. J Clin Oncol 1984; 2:961-9. 10. Parish JM, Marschke RF, Dines DE, Lee RE. Etiologic considerations in superior vena cava syndrome. Mayo Clin Proc 1981; 56:407-13. 11. Adar R, Rosenthal T, Mozes M. Vena caval obstruction: some epidemiological observations in 76 patients. Angiology 1974; 25:433-40. 12. Little AG, Golomb HM, FergusonMK, Skosey C, Skinner DB. Malignant superior vena cava obstruction reconsidered: the role of diagnostic surgical intervention. Ann Thorac Surg 1985; 40:285-8.
YELLIN, ROSEN, REICHERT, AND LIEBERMAN
1118 13. Lochridge SK, Knibbe WP, Doty DB. Obstruction of the superior vena cava. Surgery 1979; 85: 14-24. 14. Davenport D, Ferree C, Blake D, Raben M. Radiation therapy in the treatment of superior vena caval obstruction. Cancer 1978; 42:2600-3. 15. Davies PF, Shevland JE. Superior vena caval obstruction: an analysis of seventy-six cases, with comments on the safety of venography. Angiology 1985; 36:354-7. 16. YellinA, Pak HY, Burke JS, Benfield JR. Surgical management of thoracic lymphoma. Ann
Thorac Surg 1987; 44:363-9. 17. Varricchid C. Superior vena cava syndrome: an oncologic emergency. Heart Lung 1985;14:411-6. 18. Stanford W, Doty DB. The role of venography and surgery in the management of patients with superior vena cava obstruction. Ann Thorac Surg 1986; 41;158-63. 19. Dartevelle P, Chapelier A, Navajas M, et 0/. Replacement of the superior vena cava with polytetrafluorothylene grafts combined with resection of mediastinal-pulmonary malignant tumors. J Thorac Cardiovasc Surg 1987; 94:361-6.
20. Janin Y, Becker L, Wise K, Schneider D, Schwartz D, So H. Superior vena cava syndrome in childhood and adolescence: a review of the literature and report of three cases. J Pediatr Surg 1982; 17:290-5. 21. Percarpio B, Gray S. Prolonged survival following the superior vena cava syndrome. Chest 1979; 75:639-40. 22. Nogeire C, Mincer F, Botstein C. Long survival in patients with bronchogenic carcinoma complicated by superior vena caval obstruction. Chest 1979; 75:325-9.
ALON YELLIN, ADA ROSEN, NIRA REICHERT, and YAIR LIEBERMAN
Introduction Superior vena cava syndrome (SVCS) is the clinical presentation of SVC obstruction, or severe reduction in venous return from the head, neck, and upper extremities. Many mediastinal conditions may cause obstruction of the SVC either by compression or invasion, and some by thrombus formation (1). Yet for many, SVCS is almost synonymous with bronchogenic carcinoma, usually of the small cell anaplastic type. SVCS is regarded as an oncologic emergency, even in recent textbooks (2), and therapy is often initiated before a diagnosis is established. Invasive diagnostic procedures are considered hazardous by many (2-5), mainly because of excessive bleeding and respiratory complications. It was our impression that many pa-tients seen by us had conditions other than bronchogenic carcinoma. We could not recall a patient in acute distress caused by the SVCS per se. We also had to surgically obtain a tissue diagnosis in many patients in whom the SVCS was the presenting symptom of their mediastinal disease, and could not recall any outstanding complications. Bearing these controversies in mind, we reviewed our experience with SVCS during a 16-yr period. Methods During the 16 yr between 1972 and 1987, a total of 63 patients with SVCS were treated at the Sheba Medical Center, Israel. These patients were located by a computer search of the archives, and a manual search of the surgical list. All charts were retrieved, and pertinent data were recorded. Patients with venous congestion caused by cardiac failure or pericardial diseases without genuine SVC obstruction were excluded. Patients were followed until death or until early 1988 except for eight patients who could not be located. Statistical analysis consisted of a contingency table analysis using the Apple Stats Plus software. The series included 63 subjects. There was a steady increase in the annual incidence of 1114
SUMMARY This retrospective review was performed with an intent to clarify several misconceptions associated with superior vena cava syndrome (SVCS). During a 16-yr period, we diagnosed and treated 63 patients with SVCS,including seven (11%)children younger than 10 yr of age. Thirty patients (47.6%) had bronchogenic carcinoma, and 13 (20.6%) had lymphoma. In 43 cases SVCS was the presenting symptom of a mediastinal condition. Forty-one patients underwent diagnostic procedures with no major complications, and diagnosis was obtained In 36. Only six patients had surgical treatment, and 45 had radiation therapy, chemotherapy, or both. There was no mortality associated directly with venous congestion. Symptomatic relief occurred in 80% of treated patients. Weconclude that: (1)SVCSper se should not be feared (symptomatic relief is the rule), (2) accurate diagnosis can be achieved with minimal morbidity, and (3) the versatile underlying etiology dictates the outcome that can be Improved with appropriate therapy. AM REV RESPIR DIS 1990; 141:1114-1118
SVCS, though during the entire period the annual load of hospitalization had not changed. For the first period of the study (1972 to 1977)the average incidence was 1.5cases/yr. From 1978to 1983,it increased to 4.0 cases/yr, and from 1984 to 1987, to 7.5 cases/yr. There were 49 men and 14 women ranging in age from 1 to 80 yr (mean, 45 yr) (figure 1), including seven children younger than 10 yr of age. The majority were in the sixth and seventh decades of life.
Clinical Presentation The most common presenting symptoms were edema and venous congestion of the head and neck or the upper extremities followed by various respiratory complaints (table 1).Only five patients presented with hemoptysis. Chest pain, dysphagia, and speech difficulties were uncommon. None of our patients presented with, or developed, neurologic symptoms except when brain metastases were present. In 43 of 63 patients (68070), the SVCS was the presenting symptom of their disease. Their symptomatology was similar to that of patients whose SVCS developed in the course of a known disease. The duration of symptoms before admission was relatively long (table 2). In 43% of the patients, it was 5 days to 1 month, but in the remaining 34 it ranged from 1 month to 4 yr (mean, 5.5 months). Patients with malignant disease had a shorter duration of symptoms (average, 45 days) than did patients with benign conditions (average, 420 days). From the latter group, the three patients with mediastinal tumors had the longest duration
(average, 3.5 yr), whereas patients with SVC thrombosis had the shortest duration (mean, 15 days).
Underlying Conditions and Diagnostic Procedures Twenty of the patients were known to have a disease that eventually led to SVC obstruction (table 3). Of these, half had bronchogenic carcinoma, one had a lymphoma, two had primary malignant mediastinal tumors, and in three, metastatic spread from breast cancer caused the SVCS. In only four of these 20 patients was the underlying cause benign, including three children after the Mustard procedure, and one adult with SVC thrombosis after implantation of intravenous pacemaker electrodes. Of the remaining 43 patients, 41 were admitted for evaluation of SVCS, and in two the SVCS developed during hospitalization for an undetermined mediastinal condition. Bronchogenic carcinoma accounted for less than half of the cases. Lymphoma was responsible for one-fifth of all cases and for more than one-quarter of the new cases,
(Received in original form May 17, 1989 and in revised form October 17, 1989) 1 From the Department of Thoracic Surgery and Pulmonary Diseases, The Chaim Sheba Medical Center, Tel Hashomer, Israel. 2 Correspondence and requests for reprints should be addressed to Alon Yellin, M.D., Department of Thoracic Surgery, The Chaim Sheba Medical Center, Tel Hashomer 52621, Israel.
1115
THE MYTH AND FACTS OF SUPERIOR VENA CAVA SYNDROME
PTS 20
0
IS
Wales
_
Females
16 . 14 .
Fig. 1. Age distribution of patients with superior vena cava syndrome. The majority were 60 yr of age or older, corresponding to the age groups in which bronchogenic carcinoma is common.
12 10 .
0~.
JjJLD 0-10
11-20 21-30 31-40 41-50 51-60 61-70
non- Hodgkin's lymphoma underlying all but one case. Of the metastatic cases, those related to breast cancer were clinically suspected and radiologically compatible. In patients with a known malignant disease, histologic confirmation of the mediastinal mass was not sought prior to initiation of therapy. The finding of a metastatic bladder carcinoma on mediastinoscopy in one case was unexpected, as the patient was known to have a noninvasive bladder tumor. Nonmalignant conditions caused SVCS in II patients, including three cases of benign mediastinal tumors. In patients younger than 20, the condition underlying SVCS were either iatrogenic, i.e., after Mustard operation (three cases) or primary mediastinal neoplasms: lymphoma (five) and seminoma (one case). Among Arab patients, the rate of lymphoma (table 4) was significantly higher than among Jewish patients.
Diagnostic Procedures Twenty patients who had a known condition leading to SVCS were evaluated by means of TABLE 1 PRESENTING SYMPTOMS IN 63 PATIENTS WITH SVC SYNDROME Patients Symptoms
(n)
(%)
Venous congestion head and neck Swelling upper extremities or trunk Dyspnea, respiratory distress Cough, hoarseness Hemoptysis Chest pain Dysphagia Dysphonia
54 20 19 13 5 4 4
85.7 31.7 30.2 20.6 7.9 6.3 6.3 1.6
1
TABLE 2 DURATION OF SYMPTOMS BEFORE DIAGNOSIS OR THERAPY' Duration (months)
.. 1 1-6 7-12 > 12 Total
Unknown Cases
Known Diagnosis
15 25
11 5 1
3 43
17
Total
26 30 1 3 60
, n = 60; in three patients the exact duration of symptoms was unclear.
[j
>70 AGE (YRS)
conventional radiology and CT scans for anatomic localization. In the malignant cases- 16 altogether - the histologic nature of the mediastinal involvement was assumed identical to the original malignancy, and surgical procedures to prove it were usually not undertaken. In new cases, diagnosis was considered crucial (table 5). All patients had chest radiographs taken, and either linear tomography or CT scans. All but two underwent further diagnostic procedures. One elderly patient with a long-standing history was assumed to have retrosternaI goiter, and another died prior to further investigation. Semi-invasiveprocedures were helpful in 52070 of the patients,
providing histologic (seven cases), cytological (three cases), or anatomic (two cases) diagnoses. Invasive procedures ranged from lymph node biopsy (five cervical, two axillary) through mediastinoscopy to median sternotomy and formal thoracotomy. These 27 procedures were associated with no mortality and no major bleeding, obtaining tissue diagnosis in 89% of all cases. Only fivepatients, including the one with retrosternal goiter, remained without a histologic diagnosis, which proved eventually to be bronchogenic carcinoma in four.
Results
Therapy The aim of management was to alleviate symptoms and prolong life when possible in the noncurable malignant cases, treat aggressively the favorable malignant neoplasm, and achieve cure in the benign conditions. Therefore, surgery was reserved for patients with treatable vascular problems (after Mustard operation or SVC thrombosis), or benign mediastinal tumors (one goiter, one fibrous mesothelioma). Complete resection was not attempted in cases of malignant tumors in-
TABLE 3 CONDITIONS UNDERLYING SUPERIOR VENA CAVA SYNDROME (SVCS)' Newly Diagnosed Cases Tumors
(n)
(%)
Cases with Known Diagnosis (n)
20
46.5
10
Malignant Bronchogenic carcinoma Adenocarcinoma Squamous cell Small cell Large cell Anaplastic Undetermined Lymphoma Hodgkin's disease Non-Hodgkin's lymphoma Primary mediastinal tumors Thymoma Seminoma Esophageal carcinoma Metastatic tumors Breast Bladder Undetermined Total malignant
1t 1 36
Benign SIP mustard operation SVC thrombosis Retrosternal goiter Mediastinal fibrosis Behcel's syndrome Fibrous mesothelioma Total benign Grand total
2 2 1 1 1 7 43
All Cases (n)
(%)
47.6
4
2
30 6
3 4 2 3 4 12
3
6
1 11 2 1 1
4
2 1
4 27.9
1 1
4.7
2 1 1
3 3
4 8 13 2 11 4 2 1 1 4
16
52
3
3 3 2
1
16.3
4 20
6.3
6.3
3 1 1
83.7
20.6
1 1 1 11
1.6 82.5
17.5
63
, The incidence of lymphoma in previously undiagnosed cases is higher than in the entire series, and certainly higher than in patients with a known condition. t The diagnosis of bladder carcinoma was known, but the SVCS was believed unrelated.
1116
YELLIN, ROSEN, REICHERT, AND LIEBERMAN
Eighty percent of nonterminal patients achieved symptomatic improvement or complete resolution of the SVCS. In at least five, the improvement had begun before any therapy was initiated. Symptomatic relief was independent of the management. In treated patients, resolution of symptoms was unrelated to the nature of the disease. Eight of the 11 patients with benign disease (73070) achieved relief compared with 36 of the 52 (69%) patients with malignant disease (including terminal cases). With lung cancer (20 patients out of 30) the symptomatic improvement was less pronounced (11 out of 13)than in lymphoma (66% and 85%, respectively; p = NS).
TABLE 4 HISTOLOGY VERSUS ORIGIN IN SUPERIOR VENA CAVA SYNDROME"
Jew Arab Total
Cancer
Lymphoma
Other
Total
p Value
26
7 6 13
18
51 12 63
< 0.05
4
30
2
20
" Four of the six Arab patients with lymphoma were childrenyounger than 10 yr of age.
TABLE 5 PROCEDURES AIDING DIAGNOSIS IN NEW CASES OF SUPERIOR VENA CAVA SYNDROME" Condition Diagnosed Patients (n)
Diagnostic (n)
Semi-invasive Bronchoscopy Thoracentesis Venography Lymphangiography Dynamic scan Total semi-invasive
13 3 3 1 1 23
8t 2:1: 2§ 1 1§ 12
Invasive Lymph node Bx Suprasternal Bx Mediastinoscopy Mediastinotomy Median sternotomy Thoracotomy Total invasive Total procedures
7 1 15 1 2 1 27 50
511 1 14
•n
Lung Cancer
Lymphoma
6
2 2
Other
2
6
4
2
3
7
1 4
2
3 1 2
1
2 1 24 36
1
10 16
8
6
12
8
= 41.
t
Seven histology; one cytology. Both cytology. § Showed benign thrombosis secondary to mediastinal disease. I One fine needle aspiration.
*
vading the aorta or heart, nor did SVC resection or bypass seem appropriate. When the SVCS was clinically defined, all patients, except for five in terminal condition, were subjected to treatment (table 6). The type of management depended on the diagnosis. Lymphomas were treated with a curative intent, usually chemotherapy with or without radiotherapy. Patients with bronchogenic carcinomas, where the goal was to alleviate symptoms, were offered radiotherapy.
The combination of chemical agents or the amount of radiation were tailored according to the individual tissue diagnosis. Thrombolysis practiced in three patients consisted of streptokinase (two cases) and heparin infused continuously to an upper extremity vein. Medical modalities consisting of diuretics, corticosteroids, and systemic anticoagulation wereadministered to 31 patients, but were the sole treatment in two only, and therefore their efficacy cannot be evaluated.
Survival Thirty- four (54 %) patients have died (table 7), 18 during the first 6 months after the first clinical presentation, and 16patients from 6 months to 2 yr thereafter. Twenty-one (33 %) patients are alive, seven 1 yr or less after diagnosis and treatment, and 14one to seven years from the initial presentation. Patients with benign conditions had the best outcome, followed by patients with lymphoma. Twenty-five patients with bronchogenic carcinoma were followed, compared with 11 cases with lymphoma. The l-yr survival and median survival were 18%, 8.5 months, and 70%, not reached, respectively. Death in all patients was caused by the underlying disease and not by consequences of the SVCS. Discussion
Superior vena cava syndrome may result from many mediastinal conditions. It was first described by William Hunter in 1757 as a complication of a syphilitic aortic aneurysm (6). Although until the beginning of this century the most common etiologic entities associated with SVCS were benign, such as aortic aneurysms,
TABLE 6
TABLE 7
SYMPTOMATIC RESPONSE TO THERAPY
OUTCOME VERSUS ETIOLOGY
Relief Management Surgery Chemotherapy (CTx) Radiotherapy (RTx) CTx and RTx CTx, RTx, and lysis Thrombolysis Other medical None Total Terminal
Number
Followed
(n)
(%)
6 8 25 10 2 3 2 2 58 5
6 8 23 10 2 3 1 2 55
5 6 18 9 2 3 1 0
83 75 78 90
44
80
0
Alive
Etiology Benign Lung cancer Lymphoma Metastatic Mediastinal tumor Other malignancies Total
Months
Range (months)
8 5 7
51 13.2 25
13-120 7-24 4-65
21
35.3
Patients
• One was alive for at least5 months.
Dead Patients
Months
Range
3 20 4 3 3 1
7.7 10.2 8 26 5 2
0-12 0-96 2-18 13-52 1-8
34
10.4
Lost to Follow-up
5 2"
8
1117
THE MYTH AND FACTS OF SUPERIOR VENA CAVA SYNDROME
tuberculosis, and chronic mediastinitis, the pendulum has shifted towards the malignant lesions that constitute 85 to 90070 of all cases (I, 7-9). The most common malignant tumor causing SVC obstruction is bronchogenic carcinoma, which has been responsible for 67 to 82% of cases (1): neoplasms of the right lung (7) and small-cell lung cancer are more often implicated (1,9). According to Nieto and Doty (1), lymphomas underlie SVC in only 5 to 15% of the cases, and metastatic lesions in 3 to 20%. In children the most common causes are usually congenital heart disease and cardiovascular surgery, followed by lymphomas (7). In recent years there has been an increase in iatrogenic cases, i.e., those associated with insertion of pacemakers, SwanGanz catheters, and central venous lines (10). These rates are quite similar to those reported in the past from our institute by Adar and associates (11). In that study, which included patients from 1955 to 1972, there were only two patients with lymphoma. The annual incidence was 1.8 cases, which is similar to the incidence in the first period of our study. It is not entirely clear why the incidence increased progressively to 7.5 cases per year. It may be due to better awareness of this condition, more accurate recording, and a higher incidence of lung cancer. The highest proportion of patients with SVCS after the sixth decade certainly fits the age distribution of bronchogenic carcinoma. The present series includes a high proportion of patients with lymphoma in whom the SVCS heralded their disease. The modern therapy for lymphoma practically precludes the late occurrence of SVCS. Lymphoma accounted for 12% or less of the cases in other studies (8, 10, 12-15). Similar to the present experience and to our previous impression (16), Hodgkin's lymphoma was seldom a cause of SVCS. It can be argued that the high rate of lymphoma in our series may be partly due to selective referral of Arab children from the Gaza Strip. The incidence of T-Iymphoma, especially in the pediatric age group, is known to be high in that area. This probably explains why Arab patients had a much higher incidence oflymphoma than did Jewish patients, in whom the main etiology was bronchogenic carcinoma. On the other hand, it is noteworthy that SVCS is so rarely reported with malignant thymoma. In our practice thymoma (invasive and noninvasive) is more common than mediastinal lymphoma. The clinical syndrome is caused by
compression or obstruction of the SVC and altered venous return from the head, neck, and upper extremities. The severity of the SVC syndrome depends on the rapidity of occlusion and collateral vessel development - the more acute the occlusion, the more severe the syndrome (17). Collaterals drain to the azygos vein in cases of high obstruction, or to the hemiazygos or through chest wall veins to the inferior vena cava in cases of obstruction below the azygos confluence point (1). Yetin some cases the SVC may develop after surgical procedures as a consequence of destruction of mediastinal collateral vessels. Similarly, late development of collaterals may cause a spontaneous relief of symptoms without any treatment, as observed in five of our patients. In the past, and even recently, the complications of SVC obstruction, mainly laryngeal and cerebral edema, were feared (2,4). This led to the institution of emergency therapy, usually consisting of highdose radiation before tissue diagnosis was established (5, 14). Occasionally, heroic operations were performed to bypass the obstructed SVC and reduce the venous pressure (18), and even to resect what are usually considered nonoperable tumors (19). The attitude of unsound therapy also leaned on the presumed high morbidity associated with invasive procedures (2-5). Bronchoscopy should not be met with any outstanding complication unless the venous return from the lungs is also compromised. Bleeding occurring during lymph node biopsy or thoracotomy can be easily controlled. Mediastinoscopy seems hazardous, but as one is usually confronted with a limited number of dilated veins, the procedure can be accomplished by experienced surgeons with minimal complications. In our series there was only one case of minor bleeding, managed by compression. Ahmann (9) reviewed the literature and found a very low morbidity rate associated with diagnostic procedures. Recent studies (8, 12) have confirmed this observation. Nevertheless, it is justified to initiate therapy prior to precise histologic diagnosis in the occasional patient with life-threatening laryngeal or cerebral edema. Rarely, one is also confronted with a patient, usually a child, whose huge anterior mediastinal mass severely narrows the trachea (3, 20), posing a considerable anesthetic hazard. Diagnostic procedures can be delayed in such cases. It should be stressed again that the respiratory embarrassment is not due to the SVCS.
In concurrence with previous publications (7, 8, 12), we found no ground for the belief that SVCS per se frequently results in fatal complications. Moreover, not uncommonly, patients may improve spontaneously as a result of slowly enlarging collaterals. The underlying disease dictates the outcome and should therefore be managed appropriately. The span of conditions causing SVCS is wide, and management is not identical. Accurate diagnosis is crucial for planned therapy. With proper therapy a high rate of symptomatic relief can be expected, and even survival may approximate that of patients with similar diseases without SVCS (8). Long-term survivors have occasionally been reported, even in patients with bronchogenic carcinoma and SVCS (21, 22). We therefore recommend that patients presenting with superior vena cava syndrome should be carefully and promptly evaluated. Every effort should be made to obtain a precise tissue or anatomic diagnosis, and only then should specific therapy follow. Gratifying results can be expected. References 1. Nieto AF, Doty DB. Superior vena cava obstruction, clinical syndrome, etiology, and treatment. Curr Probl Surg 1986; 10:442-84. 2. Carabell SC, Goodman RL. Oncologic emergencies: superior vena cava syndrome. In: DeVita VT, Hellman S, Rosenberg SA, eds. Cancer. 2nd ed. Philadelphia: Lippincott, 1985; 1855-60. 3. Northrip DR, Bohman BK,Tsueda K. Total airway occlusion and superior vena cava syndrome in a child with an anterior mediastinal tumor. Anesth Analg 1986; 65:1079-82. 4. Salsali M, Cliff ton EE. Superior vena caval obstruction in carcinoma of lung. NY State J Med 1969; 69:2875-89. 5. Issa PY, Brihi ER, Janin Y, Slim MS. Superior vena cavasyndrome in childhood: report of ten cases and review of the literature. Pediatrics 1983; 71: 337-41. 6. Hunter W. The history of an aneurysm of the aorta with some remarks on aneurysm in general. Med Observ Inquir (Lond) 1957; 1:323. 7. Sculier JP, Feld R. Superior vena cava obstruction syndrome: recommendations for management. Cancer Treat Rev 1985; 12:209-18. 8. Schraufnagel DE, Hill R, Leech JA, Pare JAP. Superior vena caval obstruction: is it a medical emergency? Am J Med 1981; 70:1169-74. 9. Ahmann FR. A reassessment of the clinical implications of the superior vena cava syndrome. J Clin Oncol 1984; 2:961-9. 10. Parish JM, Marschke RF, Dines DE, Lee RE. Etiologic considerations in superior vena cava syndrome. Mayo Clin Proc 1981; 56:407-13. 11. Adar R, Rosenthal T, Mozes M. Vena caval obstruction: some epidemiological observations in 76 patients. Angiology 1974; 25:433-40. 12. Little AG, Golomb HM, FergusonMK, Skosey C, Skinner DB. Malignant superior vena cava obstruction reconsidered: the role of diagnostic surgical intervention. Ann Thorac Surg 1985; 40:285-8.
YELLIN, ROSEN, REICHERT, AND LIEBERMAN
1118 13. Lochridge SK, Knibbe WP, Doty DB. Obstruction of the superior vena cava. Surgery 1979; 85: 14-24. 14. Davenport D, Ferree C, Blake D, Raben M. Radiation therapy in the treatment of superior vena caval obstruction. Cancer 1978; 42:2600-3. 15. Davies PF, Shevland JE. Superior vena caval obstruction: an analysis of seventy-six cases, with comments on the safety of venography. Angiology 1985; 36:354-7. 16. YellinA, Pak HY, Burke JS, Benfield JR. Surgical management of thoracic lymphoma. Ann
Thorac Surg 1987; 44:363-9. 17. Varricchid C. Superior vena cava syndrome: an oncologic emergency. Heart Lung 1985;14:411-6. 18. Stanford W, Doty DB. The role of venography and surgery in the management of patients with superior vena cava obstruction. Ann Thorac Surg 1986; 41;158-63. 19. Dartevelle P, Chapelier A, Navajas M, et 0/. Replacement of the superior vena cava with polytetrafluorothylene grafts combined with resection of mediastinal-pulmonary malignant tumors. J Thorac Cardiovasc Surg 1987; 94:361-6.
20. Janin Y, Becker L, Wise K, Schneider D, Schwartz D, So H. Superior vena cava syndrome in childhood and adolescence: a review of the literature and report of three cases. J Pediatr Surg 1982; 17:290-5. 21. Percarpio B, Gray S. Prolonged survival following the superior vena cava syndrome. Chest 1979; 75:639-40. 22. Nogeire C, Mincer F, Botstein C. Long survival in patients with bronchogenic carcinoma complicated by superior vena caval obstruction. Chest 1979; 75:325-9.
