A Comparison of Carbon Dioxide and Radiopaque Angiocardiographic Methods in the Diagnosis of Pericardial Effusion

Abstract

Diagnostic accuracy in pericardial effusion is of utmost clinical importance if the aspiration of fluid from the pericardial sac is contemplated. Unfortunately, the findings of conventional roentgenograms, even in the presence of massive effusion, are frequently equivocal. During the last two decades the proper recognition of fluid in the pericardium has been rendered more accurate with the aid of angiocardiography, utilizing either radiopaque substances (5, 7, 9, 14, 16) or carbon dioxide as the contrast medium (3, 11–13). The application of angiocardiography with radiopaque contrast medium to the differential diagnosis of pericardial effusion and cardiac dilatation was first emphasized by Williams and Steinberg (16). In 1957 Figley and Bagshaw (5) discussed the angiocardiographic aspects of constrictive pericarditis and established criteria important in the differential diagnosis of pericardial thickening and effusion. Later, Durant et al. (3) introduced a new method of angiocardiography specifically for studying pericardial disease by outlining the lateral border of the right atrium with intracardiac carbon dioxide. Subsequently, Scatliff and his associates (12) and Stauffer, et al. (13) discussed the value of this procedure in elucidating the diagnostic problems of pericardial disease. While the above studies attest to the value of these procedures in the differentiation of pericardial fluid from cardiac enlargement, little information is available with respect to the limitations and relative merits of these methods when performed in the same patient. For several years we have employed angiocardiography with both radiopaque substances and carbon dioxide in pericardial effusion and have been impressed that the two media occasionally yield different information in the same patient. It is the purpose of this study to show that there are discrepancies between the two contrast methods and that certain pitfalls may be encountered which have important clinical implications. Our findings also provide information regarding shifting of pericardial fluid with changes in the position of the patient. Technic and Method The roentgenograms of 31 patients with suspected pericardial effusion studied by both radiopaque and carbon dioxide angiocardiography were available for analysis. Each examination was performed in the following manner: radiopaque angiocardiography was carried out first with 40–50 cc of contrast medium, the patient being in the right lateral decubitus position (right side down, horizontal x-ray beam). The patient was immediately turned into the left decubitus position (right side up) and 50–100 cc of carbon dioxide was injected. The patient was kept in this position for five to ten minutes while the roentgenograms were developed.