Differential Diagnosis

The etiology of a pericardial effusion in a cancer patient may not always be apparent. It can result from various factors, and noninvasive evaluation may not always be sufficient to establish the etiology. The patient's history is usually very helpful, and information regarding previous irradiation of the chest or previous nonspecific pericarditis with effusion is especially pertinent. High fever, leukocytosis, and prostration suggest septic pericarditis, whereas recent myocardial infarction could suggest an association with the ischemic event. When the cause of a pericardial effusion remains unclear, diagnostic (cytologic, bacteriologic, chemical, and immunologic) analysis of a fluid specimen may be useful. In some instances, the etiology of pericardial effusion in a cancer patient remains elusive despite an aggressive attempt at evaluation; even pericardial biopsy does not always provide a definitive explanation. Interestingly, the serum cancer antigen 125 (CA-125) level may be elevated in patients with pericardial and pleural effusions, as well as in patients with heart failure, even in the absence of a gynecologic malignancy;^27,28 conditions that irritate or inflame serosal surfaces appear to induce apparently normal mesothelial cells to express this antigen.

Cardiac Tamponade

The accumulation of pericardial fluid may increase intrapericardiac pressure and compromise cardiac output, a condition known as cardiac tamponade. Symptoms include dyspnea and exercise intolerance, and signs include tachycardia, neck-vein distention, and hepatomegaly.^{29, 30} Heart sounds are often, but not always, difficult to auscultate, and pericardial friction rubs may or may not be present.³¹ Vague chest discomfort or fullness is frequently noted. Most patients with significantly increased intrapericardiac pressure also demonstrate an exaggeration of the decrease in pulse pressure during inspiration (pulsus paradoxus). The Kussmaul sign (distention of the jugular veins on inspiration) is seen more commonly in the context of constrictive pericarditis or mediastinal tumors. A highly characteristic finding of cardiac tamponade is electrical alternans, in which the electrocardiographic QRS voltage becomes larger and smaller on alternate complexes. This phenomenon is caused by physical movement of the heart toward and away from the electrode as the heart rocks back and forth within the fluid-containing pericardial sac. Cardiac tamponade can almost always be diagnosed on the basis of the physical findings and the findings from noninvasive studies. Echocardiography usually shows the effusion, which may vary considerably in size. Diastolic inward motion (collapse) of the right atrium is a sensitive but not specific finding; the finding is more specific when the collapse is greater than one-third of the cardiac cycle. Diastolic collapse of the right ventricular wall becomes more pronounced as hemodynamic compromise progresses and is a much more specific finding. Doppler flow studies show exaggerated tricuspid and pulmonic flow velocities and reduced transmitral flow velocity with the onset of inspiration. Cardiac catheterization may show a graphic representation of pulsus paradoxus, and there may be elevation and ultimate equalization of the diastolic pressures in the cardiac chambers as tamponade progresses. When frank tamponade ensues, the pulse becomes weak or totally absent during inspiration, and patients develop symptoms of low-output cardiogenic shock. Death, often preceded by profound bradycardia, ensues if the tamponade is not promptly resolved.

Management of Malignant Pericardial Effusions and Tamponade

The management of malignant pericardial effusion depends on a number of factors, including the likelihood of the tumor responding to local (surgical, radiotherapeutic, or intracavitary) or systemic therapy; the extent of and the symptoms attributable to the effusion; and the overall anticipated survival of the patient.³² Patients with tumors highly likely to respond to the systemic therapy may proceed with their treatment; sometimes the malignant effusion resolves in responses to the systemic anticancer therapy alone. Pericardial effusion in patients with tumors that are unlikely to respond to treatment, either because of the cell type or because the tumor has become refractory to treatment, may require a local intervention. In patients who have a more favorable oncologic prognosis and are sufficiently strong to undergo general anesthesia and surgery, creation of a pleuropericardial window is often considered the procedure of choice.³³ The communication formed usually remains patent, and the larger surface area available in the pleural space permits more effective resorption of the excess fluid. The procedure is generally well tolerated, but pericardial needle drainage may be required prior to surgery as a stabilizing measure in patients who have very large effusions or pretamponade compromise of cardiac output. Symptoms often resolve dramatically after removal of the fluid, allowing patients to again engage in activities that had become impossible for them. Needle drainage is now almost always undertaken with echocardiographic confirmation of the position of the draining catheter and is considered safe and effective.^{34, 35} Some patients experience transient left ventricular dysfunction after resolution of cardiac tamponade, and thus a period of careful monitoring is important.³⁶ The advantages of the pericardial window over percutaneous pericardiocentesis have not been fully evaluated, but some studies suggest little advantage to the more invasive pericardial window procedure. The advisability of routine drainage of large pericardial effusions in patients without tamponade has also been questioned; these authors point out the low diagnostic yield and the lack of therapeutic benefit.³⁷ The clinical management of such patients should be determined by their overall performance status and oncologic prognosis and by the expertise at the treatment center.³²

Sclerosis may be considered following needle drainage. A number of sclerosing agents have been studied. These include hyperosmolar glucose, radioactive gold, bleomycin, sterile talc, doxycycline (Vibramycin), and triethylenethiophosphoramide (thio-TEPA).^38,39 Doxycycline a semisynthetic tetracycline used at a dose of 250 to 500 mg, has received considerable attention as an effective sclerosing agent.⁴⁰ Sclerosis is carried out by instilling the sclerosing agent directly into the pericardial space; the procedure may be repeated every 24 to 48 hours until the pericardial fluid reaccumulation falls below 50 mL per day. For most patients, sclerosis of the pericardial sac will occur after three to five instillations. The likelihood of pericardial sclerosis succeeding in a specific patient is difficult to predict, but historically the procedure is helpful in more than half the patients.⁴¹ It is clearly a useful adjunct, since it may obviate repeated pericardiocentesis and prevent recurrent cardiac tamponade. The procedure is usually well tolerated, although some patients experience considerable pain and discomfort that may require narcotic analgesia.

Ziskind and colleagues have described using percutaneous pericardiotomy to relieve pericardial effusions.⁴² In this technique, a balloon catheter is inserted so that the balloon crosses the pericardium. The balloon is then inflated, which tears the pericardium open, allowing the fluid to drain. The procedure may be quite painful, and patients require sedation and analgesia or anesthesia. Nonetheless, these authors reported success in 46 of 50 patients. Additional reports confirm that balloon intervention is a safe alternative to the creation of a pleuro-pericardial window.^43,44