[Diagnosis of interatrial communications using contrast echocardiography]

M-mode contrast echocardiography with peripheral venous injections was performed in 73 patients with interatrial communications: 48 (group 1) had a hemodynamically significant atrial septal defect (ASD), 19 (group 2) had a patent foramen ovale (PFO) without clinical or oxymetric evidence of a shunt. The remaining 6 (group 3) had an interatrial communication in combination with severe additional congenital malformations predisposing to a right to left (R-L) shunt. Contrast studies were considered positive for a shunt lesion when at least five clearly recognizable contrast echoes appeared in the left heart following one injection. During quiet respiration positive contrast studies were obtained in 85% of all ASD patients (including all 10 with Eisenmenger's reaction and 31/38 [82%] uncomplicated cases); in 37% of the PFO cases (including 3/13 with normal right heart pressures), and in 53/73 (73%) of all patients with interatrial communications. The intensity of contrast shunting was variable in all groups. Opacification of the mitral funnel (which is typical for an atrial level shunt) was observed in 45 patients. In 8 patients with positive studies the few contrast echoes, that appeared in the left heart were first seen after they had left the mitral valve. Contrast injections into the pulmonary artery were performed in a control group of 29 patients. No contrast appeared in the left heart as expected. In 57 patients (39 ASD, 17 PFO, 1 group 3) contrast studies were also performed during the Valsalva maneuver. Valsalva provocation resulted in increased contrast shunting in 19, led to new mitral funnel opacification in 9 and improved the sensitivity of contrast echocardiography by 9 and 26% in ASD and PFO cases respectively. The intensity of contrast shunting was largely independent of the hemodynamic findings and was often variable upon subsequent injections in the same patient. Therefore contrast echocardiography is not helpful in predicting the L-R shunt or the pulmonary artery pressure and does not seem suited for follow-up studies. The differentiation between true contrast echoes in the left heart and artifacts, noise echoes, "overload", or incomplete mitral structures and the differentiation between interatrial and interventricular contrast shunting is usually easy. However the distinction between a hemodynamically significant ASD and pulmonary arteriovenous fistulas, certain venous anomalies or a patent foramen ovale may be difficult or even impossible by contrast echocardiographic criteria alone. Resting two-dimensional contrast echocardiograms were recorded in 57 patients including 34 with ASD, 18 with PFO and 5 from group 3.(ABSTRACT TRUNCATED AT 400 WORDS)