Detection of pericardial effusion by chest roentgenography and electrocardiography versus echocardiography.

To determine the sensitivity and specificity of chest roentgenography and electrocardiography in the detection of pericardial effusion, echocardiography was used as the diagnostic standard. Chest roentgenograms and electrocardiograms of 124 patients, 57 of whom had pericardial effusion, were read without knowledge of the echocardiographic interpretation. The sensitivity of roentgenographic diagnosis was low (20%), as was that of diagnosis from decreased voltage on the electrocardiogram (26%). The specificity of the chest roentgenogram was 89% and that of the low-voltage electrocardiogram 97%. The high specificity of the low-voltage electrocardiogram may have been due in part to the exclusion of obese and emphysematous subjects from the study. When cardiomegaly detected roentgenographically or a low-voltage electrocardiogram or both were considered as evidence of pericardial effusion, sensitivity improved to 82% but specificity declined to 29%. It is concluded the chest roentgenography and electrocardiography are unsatisfactory as screening investigations for the detection of pericardial effusion.