The superior vena cava (SVC) syndrome develops as a result of impaired blood return through the SVC to the right atrium. Obstruction results in venous hypertension, with the severity of ensuing signs and symptoms dependent on the site of obstruction and the rapidity with which the block occurs. The SVC is formed by the union of the left and right brachiocephalic veins in the middle third of the mediastinum and extends inferiorly
for 5 to 8 cm, terminating in the right atrium (Fig. 116.1). The SVC serves as the principal venous drainage for the head, neck, and upper extremities. The major collateral, the azygous vein, joins posteriorly just over the right mainstem bronchus and drains the posterior thorax. The SVC is thin walled and is bounded by the mediastinal parietal pleura and the right paratracheal, azygous, hilar, and subcarinal lymph nodes. As a result, it is extremely susceptible to extrinsic compression by adjacent lymph nodes or the aorta, with subsequent stasis, occlusion, or thrombosis. If obstruction occurs distal to the azygous vein, collateral flow through the azygous can adequately compensate for diminished return. However, if the obstruction is proximal to the azygous, flow must completely bypass the SVC and return via internal mammary, superficial thoracoabdominal, and vertebral venous systems to the inferior vena cava. This more circuitous route results in significantly higher venous pressures. The trachea and bronchi of children are smaller and significantly more susceptible to extrinsic compression, increasing the risk of fatal complications.

The vast majority of patients with SVC syndrome have bronchogenic carcinoma, most commonly of the small cell histology (Table 116.1). Non-Hodgkin lymphoma, breast cancer, and other neoplasms make up the remainder of the malignant causes. Despite a high frequency of mediastinal involvement, Hodgkin lymphoma patients rarely present with SVC compression. Benign causes of SVC syndrome make up 6% to 20% of all cases and include thrombosis due to indwelling intravenous catheters or pacemakers and granulomatous disease [1,2]. Infectious causes of the SVC syndrome decreased substantially with the advent of antibiotics but must be considered in the differential diagnosis for patients from endemic areas or with potential human immunodeficiency virus infection. Blastomycosis, actinomycosis, histoplasmosis, tuberculosis, nocardia, and syphilis occasionally cause fibrosing mediastinitis and aortitis leading to SVC syndrome. An extensive list of rare causes may include idiopathic mediastinal fibrosis, goiter, thymoma, Behçet’s syndrome, sarcoidosis, prior radiation with local vascular fibrosis, and unusual metastases of common malignancies.

The presentation of SVC syndrome depends largely on the acuity of the obstruction to flow. In patients with benign causes, extensive collateral flow often develops that minimizes symptoms
for months to years. Acute compression by tumor or thrombosis does not allow time for collateralization, and venous hypertension inevitably induces symptoms. Symptoms include dyspnea, edema of the face, neck, upper torso, and extremities; and cough. In rare instances, patients complain of hoarseness, syncope, headaches, chest pain, or dysphagia due to esophageal compression. Physical signs include jugular venous distention, edema of the face or upper extremities, dilated venous collaterals, plethora, and tachypnea and, in rare instances, papilledema or stridor.

Initial evaluation should include a chest radiograph and contrast enhanced computed tomography (CT) to confirm the clinical diagnosis, identify a potential etiology, and localize the obstruction. Venography or magnetic resonance imaging (MRI) may be appropriate in subsequent evaluation to better define the extent of obstruction, particularly if stenting of the obstruction is considered (Fig. 116.2). Of note, focal hepatic contrast enhancement on CT has been noted in patients with SVC syndrome due to collateralization through patent remnants of the umbilical vein or of the musculophrenic venous system [3]. These abnormalities could be mistaken for metastatic disease and should be further evaluated in patients in whom therapy would be changed in the presence of isolated metastases. If a malignant cause of SVC obstruction is considered, all reasonable efforts should be made to obtain diagnostic material, as treatment depends on the underlying histology. The approach may include sputum cytology, bronchoscopy, transthoracic needle aspiration, biopsy of palpable lymph nodes, mediastinoscopy, thoracotomy, or video-assisted thoracoscopy. Despite concerns regarding surgical complications, morbidity associated with surgical procedures necessary to procure a diagnosis is not substantially different from that in patients without SVC syndrome [1,4]. The rapidity of the diagnostic workup depends on the likelihood of morbid complications at the time of presentation. Most series suggest that patients with malignant SVC syndrome have had symptoms an average of 45 days before presentation, and the vast majority of patients with malignancy do not die of SVC syndrome but of other complications of their disease [1]. The significant complications of SVC syndrome are tracheal obstruction and cerebral edema, and fatalities from cerebral edema are extremely rare. Therefore, the truly emergent situation in adults is a patient who presents with stridor or other evidence of significant airway compromise or the rare patient with cerebral edema. In essentially all other settings, treatment should be instituted only after the malignant or benign cause of the syndrome has been established because outcome is not compromised by delay for appropriate evaluation.

Once the diagnosis is established, initiation of therapy depends on the etiology, the severity of symptoms, the acuity of presentation, and the goals of treatment. If patients are minimally symptomatic, the azygous is patent, and treatment is focused on palliation, observation is a reasonable option. Chemotherapy is the treatment of choice for SVC syndrome due to small cell lung carcinoma, non-Hodgkin lymphoma, and germ cell tumors. Although radiation is often considered in addition to chemotherapy even in the palliative setting, 80% of these patients have a complete or partial response of their symptoms to chemotherapy alone [5]. Other histologies should be treated with endovascular stent placement, radiation therapy, or both. Radiation therapy prior to biopsy has been associated with a significant reduction in rates of histologic diagnosis and should be avoided [6]. Although external beam radiation effectively palliates symptoms in more than 70% of patients within 2 weeks [7], relapse after radiotherapy occurs in 15% to 30% of cases.

Endovascular stent placement, as a primary intervention, is a particularly attractive option for patients who lack a tissue diagnosis and whose symptoms on presentation require a rapid palliative intervention; including all patients, regardless of histology, who present with airway compromise or cerebral edema. In these patients, SVC stent placement provides rapid relief of symptoms (less than 48 hours) while awaiting response to systemic chemotherapy or radiation treatment. Responses to endovascular stent placement are durable (90% symptom free at time of death, versus 12% with palliative radiation) and the primary patency rates for malignant SVC syndrome are 50% to 100% [8]. Some authors have suggested a role for stent placement in first-line management of all SVC syndrome patients; however, no randomized controlled trials have been published [8].

Anticoagulation after stent placement is controversial, with some studies suggesting a high rate of thrombosis unless patients are anticoagulated, whereas other series, using no anticoagulation, report efficacy and thrombotic risk equivalent to those who use anticoagulation [9,10].

In patients with an established diagnosis, radiation therapy remains an appropriate intervention. Many fractionation protocols have been used, with the majority of patients receiving 30 Gy in 10 fractions, whereas patients treated with curative intent often receive 50 Gy in 25 fractions. Although high doses of radiation have often been given early in the treatment course to achieve rapid tumor response, there is little evidence to suggest that this is necessary [11]. In cases of SVC thrombosis with an indwelling catheter or pacemaker, thrombolytic agents may be useful as primary therapy or as an adjunct to stent placement [12]. The additional benefit of thrombolytics or anticoagulation in patients treated for malignant SVC syndrome is not well established.

Surgical resection and reconstruction of the SVC is reserved for patients with benign disease or the rare patient with tracheal obstruction in the setting of chemotherapy or radiotherapy-resistant disease.

SVC syndrome has been thought to predict for poor outcome. However, the presence of SVC syndrome is not a negative prognostic factor in small cell carcinoma and lymphoma independent of the stage and bulk of disease, and patients should be treated with curative intent if otherwise appropriate [13].

Cardiac tamponade results from accumulation of fluid within the pericardium that impairs left ventricular expansion and diastolic filling. As stroke volume drops, compensatory tachycardia occurs to offset progressive hypotension. Ultimately, pressures equalize in the left atrium, pulmonary vasculature, right atrium, and SVC, and circulatory collapse ensues. As with SVC syndrome, the severity of symptoms is dependent on the speed of progression. Tamponade occurs when the pericardium cannot expand because fluid accumulation is too rapid or because the pericardium is thickened or fibrotic.

Pericardial or cardiac involvement with malignancy occurs in 1% to 20% of patients with cancer and is often not diagnosed antemortem [14,15]. In up to 40% of unselected patients presenting with tamponade, malignancy is identified as the cause; the frequency with which tamponade develops as the initial manifestation of a patient’s disease has led to standard cytologic examination of all significant effusions [16]. Tumors may involve the pericardium by direct extension from intrathoracic organs or hematogenous spread. Malignancies most often associated with pericardial effusions are lung, breast, lymphoma, and leukemia. Pericardial effusions in patients with cancer are due to pericardial or cardiac involvement in 60% of cases, with idiopathic pericarditis and radiation-induced pericarditis causing 32% and 10% of cases, respectively [17]. Other potential causes include infection, Dressler’s syndrome, rheumatic disease, and hypothyroidism.

The common symptoms of pericardial effusion include dyspnea (85%), cough (30%), orthopnea (25%), and chest pain (20%). The common signs of pericardial effusion are jugular venous distention (100%), tachycardia (100%), pulsus paradoxus (89%), systolic blood pressure of less than 90 (52%), and pericardial rub (22%) [16]. Other signs of right- and left-sided heart failure may include hepatosplenomegaly, rales, peripheral edema, and ascites. Plain films demonstrate cardiac enlargement in at least half of all cases, and electrocardiography may reveal abnormalities suggestive of pericarditis (low-voltage, ST-segment elevation) or electrical alternans.

Echocardiography is the most useful means of rapidly detecting hemodynamically significant effusions. Early signs include right atrial collapse and mitral regurgitation with later detection of left atrial or right ventricular collapse. Echocardiography also allows estimation of the volume, fluidity, and contents of the effusion, although it is difficult to distinguish tumor, thrombus, or fibrinous material from one another. The specificity of echocardiography for hemodynamic compromise has been called into question, and in many centers right heart catheterization with demonstration of equalization of pressures is required to diagnose tamponade physiology definitively. Emergent treatment of tamponade invariably involves drainage of the effusion, and cytologic evaluation of the fluid provides a very specific means of establishing a malignant etiology. The detection rate of pericardial fluid cytology ranges from 50% to 100%, and certain histologies, such as lymphoma and mesothelioma, are more difficult to demonstrate in pericardial fluid [14,18,19]. Pericardial biopsy is occasionally required to establish a diagnosis in difficult cases and can be performed under local anesthesia using a subxiphoid approach. The presence of a pericardial effusion correlates with a shortened survival among patients with cancer (median survival 15.1 weeks) and the definitive identification of neoplastic cells in the pericardial fluid by cytology portends an even worse prognosis (median survival 7.3 weeks) [20].

Cardiac tamponade requires immediate treatment to relieve the increased end-diastolic pressure and inadequate ventricular filling. Oxygen, pressor agents, and intravenous fluids to improve cardiac output should be provided as appropriate. Inotropic agents are frequently ineffective however, because a state of intense adrenergic stimulation is already present [21]. When airway management is required, significant caution should be used because the positive intrathoracic pressure that results from initiation of mechanical ventilation places tamponade patients at particularly high risk for profound postintubation hypotension [21]. Emergent pericardiocentesis is indicated for significant hypotension, and it has been suggested that a pulse pressure of less than 20 mm Hg, a paradoxic pulse greater than 50% of the pulse pressure, or a peripheral venous pressure above 13 mm are other absolute indications for emergent intervention [22]. Fluid should be evaluated with cell counts, cultures, and cytology as noted earlier. Patients who present with malignant tamponade have recurrence after simple pericardiocentesis in 58% to 83% of cases [16,23]. Pericardial effusions without clinical tamponade may be observed if patients are asymptomatic or have minimal effusion (less than 1 cm), as progression to tamponade requiring pericardiocentesis in a single study was 20% for all patients, and progression of effusions of less than 1 cm in size to greater than 1 cm was only 4% [23]. Because of the high recurrence rate after pericardiocentesis in patients with tamponade, additional therapy is generally indicated if the patient’s survival or quality of life would be otherwise compromised. Symptomatic relief with pericardiocentesis alone is 90% to 100%, with a complication rate of 3% [24]. Radiation therapy is noninvasive and allows treatment of the majority of the pericardium but carries a theoretical risk of radiation-induced pericarditis. As a single modality, radiation controls pericardial effusion in 67% of cases, with a particularly high success in hematopoietic tumors (93%). Systemic therapy is generally used only for diseases that are considered to be chemosensitive, such as breast cancer or lymphoma; in these individuals, it prevents recurrence in 73% of treated patients.

Instillation of sclerosing agents, radionuclides, and chemotherapy through indwelling catheters have been widely used with the intent to induce nonspecific inflammation with obliteration of the pericardial space or to achieve specific antineoplastic effects. Typically, a catheter is placed into the pericardial sac and drainage continued until output is less than 100 mL per day. Sclerosing agent or chemotherapy is injected into the catheter every 24 to 48 hours until fluid output is less than 25 to 50 mL per day, and the catheter is removed. A review of 20 different studies reported an overall control rate of 82% with common toxicities, including fever, pain, arrhythmias, and occasional cytopenias [24]. Tetracycline, which is no longer available, has the most extensive track record; however, doxycycline and minocycline have shown similar efficacy in malignant pericardial and pleural effusions. Chemotherapeutic agents that demonstrate response rates greater than 50%
include bleomycin, cisplatin, carboplatin, mitoxantrone, fluorouracil, and thiotepa [24,25,26,27,28,29]. In a randomized trial of 80 patients comparing intrapericardial bleomycin with observation alone following drainage, the 2-month failure free survival was 46% versus 29%; and median survival was 119 days versus 79 days for the groups, respectively. Because of the small size of this trial, these differences did not achieve statistical significance [30].

One small prospective trial (n = 21) comparing bleomycin with doxycycline showed bleomycin to be better tolerated, with less retrosternal pain and shorter periods of catheter drainage [28]. The use of sclerosing agents in the treatment of recurrent malignant pericardial effusions may result in an increased risk of both subsequent constrictive pericarditis and tamponade, leading some groups to favor the instillation of nonsclerosing chemotherapeutic agents [31]. In the absence of randomized studies, no single agent is accepted as the gold standard for intervention.

A surgical procedure or balloon catheter can be used to create a pericardial window to drain the fluid. This can be done by performing a subxiphoid pericardiotomy, thoracotomy or thoracoscopy with window, pleuroperitoneal window, or subcutaneous balloon pericardiotomy. These procedures control the effusion in 85% to 95% of patients [24,32,33,34]. An advantage of subxiphoid or balloon pericardiotomy is that both can be performed without general anesthesia, reducing operative morbidity.

The development of malignant pericardial effusion and tamponade usually reflects uncontrolled metastatic disease and portends a dire prognosis. Median survivals for patients treated for tamponade range from 3.3 to 4.5 months. Nonrandomized studies suggest that patients with lung and breast cancer have substantially better survival rates if systemic therapy can be instituted [35,36]. The decision to intervene in a patient with malignant cardiac tamponade depends on the patient’s histology and sensitivity to treatment as well as the patient’s condition. Patients for whom treatment of tamponade provides meaningful palliative benefit should be considered for the treatment that is likely to provide durable relief of symptoms with the minimum of morbidity and requirement for hospitalization.

Epidural cord compression by malignancy occurs as a result of metastasis or primary tumor involvement of the vertebral column, paravertebral space, or epidural space. Damage to the cord occurs when the tumor compromises the vertebral venous plexus or compresses neural tissue directly or when compromised bone impinges on the cord. The resulting vasogenic edema and hemorrhage induce further ischemic damage. The vertebral body is the most common source of compressive lesions, predominantly in the thoracic (70%), followed by the lumbar (20%) and cervical (10%) regions [38]. Tumor invasion through the intervertebral foramen and cord compression without bone involvement is most often seen with lymphoma, leading to normal plain films and radionuclide scans despite clinical compression. Multiple noncontiguous levels are involved in 10% to 40% of cases [39,40].

The most common causes of malignant cord compression are tumors with a propensity for bony metastases, including breast and lung, followed by hematopoietic malignancy and gastrointestinal and genitourinary primaries [41,42] (Table 116.2). Cord compression afflicts 5% of patients during their course and is found in up to 10% of patients at autopsy. Benign causes of cord compression include stenosis, epidural abscess, or hematoma.