Respiratory changes in transvalvular flow velocities versus two-dimensional echocardiographic findings in the diagnosis of cardiac tamponade

Background. The purpose of this study was to compare the sensitivity, specificity and positive predictive value of the respiratory changes in the transvalvular flow velocities to those of right atrial collapse and right ventricular collapse in the diagnosis of cardiac tamponade. Methods. Standard two-dimensional and Doppler echocardiography were performed with respiratory monitoring in 56 consecutive patients with mild to severe pericardial effusion. Sixteen patients met the clinical criteria for cardiac tamponade and underwent pericardiocentesis or surgical drainage. Forty patients were found to have no tamponade and were followed up for at least 2 weeks and none of them showed clinical worsening. Results. The sensitivity, specificity and predictive value were, respectively, 77, 80 and 62% for an inspiratory decrease > 22% in the peak velocity of the early mitral flow; 75, 89 and 73% for an inspiratory reduction > 20% in the peak velocity of the aortic flow; 50, 69 and 36% for an inspiratory increase > 30% in the peak velocity of the early tricuspid flow; 87, 85 and 64% for an inspiratory increase > 25% in the peak velocity of the pulmonary flow. Right atrial collapse and right ventricular collapse had a sensitivity of 100 and 75%, a specificity of 33 and 85%, and a predictive value of 37 and 66%, respectively. Conclusions. In the diagnosis of cardiac tamponade: 1) right atrial collapse is the most sensitive sign but lacks any specificity; 2) except for the tricuspid valve, the respiratory variations in the transvalvular flow velocities have a reliability and a predictive value comparable with those of right ventricular collapse; 3) the predictive value is not very high, indicating that at both techniques false positive results are not negligible.