Comparison of contrast echocardiography versus cardiac catheterization for detection of pulmonary arteriovenous malformations.

Because clinical diagnosis of pulmonary arteriovenous malformations (AVMs) is difficult and cardiac catheterization invasive, contrast echo has been used to aid in the diagnosis. Correlation between contrast echo and catheterization data in these patients remains poorly defined. We examined the ability to detect pulmonary AVMs by contrast echocardiography, pulmonary angiography, and pulmonary vein saturations in patients who have undergone cavopulmonary (Glenn) shunts. Pulmonary vein saturations were obtained from catheterization reports. Blinded observers reviewed angiographic and contrast echo data retrospectively in 27 patients who had undergone a Glenn shunt and in 19 controls with biventricular hearts. Contrast echo was positive in 68 of 99 lungs (69%). Angiography showed AVMs in 65 of 98 lungs (66%). Pulmonary vein desaturation was found in 13 of 45 lungs sampled (29%). Only 10 of 38 lungs with positive contrast echo had pulmonary vein desaturation, but only 1 patient had pulmonary vein desaturation without positive contrast echo. The degree of desaturation did not correlate with severity of contrast echo return; 6 of 11 patients with 3+ contrast echo studies had normal pulmonary vein saturations. In a control group of patients with biventricular hearts, only 1 of 19 (5.3%) contrast echos was weakly positive. Contrast echo correlates poorly with angiography and pulmonary vein saturation for evaluation of AVMs. Contrast echo is extremely sensitive and often positive despite normal pulmonary vein saturation. Assessment of pulmonary vein desaturation in 100% oxygen may improve its sensitivity and correlation with contrast echo. Additional studies are needed to follow patients with mildly positive contrast echo studies to enable determination of their clinical relevance.