Echocardiographic Assessment of the Accuracy of Computed Tomography in the Diagnosis of Hemodynamically Significant Pericardial Effusions

Background: Although Doppler echocardiography is the noninvasive test of choice for the diagnosis of hemodynamically significant pericardial effusion (HSPE), pericardial effusion is often first detected by computed tomography (CT). Few data are available, however, regarding the accuracy of CT in the diagnosis of cardiac tamponade. Methods: We retrospectively identified 28 consecutive hospitalized patients with pericardial effusion who underwent both chest CT and echocardiography within 48 hours between 2001 and 2005. CT images were analyzed by experienced radiologists without knowledge of the echo results. Radiologic findings were evaluated with Doppler echocardiography. Results: Of the 28 patients, 15 had HSPE by echo criteria. Right ventricular wall flattening and pericardial thickening on CT correlated with HSPE with a positive predictive value of 79% and 67%, respectively. Inferior vena cava diameter by CT was not a useful predictor of HSPE. CT was most accurate in diagnosing HSPE when the effusions were large (n = 6) and was 100% sensitive and specific in this group. Of the small- to moderate-sized effusions (n = 22), the sensitivity and specificity were 80% and 75%, respectively. Conclusions: CT findings of right ventricular wall flattening, pericardial thickening, and a large pericardial effusion significantly increase the likelihood of tamponade. In this study, the majority of patients diagnosed with HSPE by CT did indeed have tamponade by standard echo criteria. Although not a useful screening tool in the diagnosis of HSPE, CT is of some help, and abnormal results warrant more urgent clinical evaluation.