BackgroundBecause pulmonary artery size is considered by most investigators to be a major prognosticator of outcome in patients undergoing staged Fontan reconstruction, the objective of the present study was to determine the efficacy of noninvasive measures in determining pulmonary artery size.Methods and ResultsThis study analyzed the T1-weighted, spin-echo magnetic resonance and echocardiographic images of 36 functional single-ventricle patients throughout stages of Fontan reconstruction (prebidirectional and postbidirectional cavopulmonary anastomosis and after Fontan) and compared them with angiography images at cardiac catheterization. Magnetic resonance imaging had a high degree of agreement with angiography, with the McGoon index agreeing better than the Nakata index and absolute right and left pulmonary diameters. Although echocardiography had fair agreement with angiography, it agreed less well and had a wider standard deviation than magnetic resonance imaging for all indexes and measurements and, based on the prediction interval, would be a poorer prospective measure of pulmonary artery size in this population. In addition, echocardiography was a poorer measure of pulmonary artery size as the size of the vessel increases. Magnetic resonance imaging correctly detected five of five patients with nonconfluent branch pulmonary arteries and six of six patients with stenoses, whereas echocardiography was unable to visualize any of the patients with nonconfluent branch pulmonary arteries with certainty and only two of six (33%) with stenoses.ConclusionsMagnetic resonance imaging is a useful, noninvasive tool to determine pulmonary artery size in patients undergoing Fontan reconstruction and is superior to echocardiography. Echocardiography was a fair predictor of pulmonary artery size, but magnetic resonance imaging agreed with angiography better than echocardiography and outperformed echocardiography in diagnosing branch pulmonary artery discontinuity and stenoses. Magnetic resonance imaging may avoid unnecessary cardiac catheterization, especially in older patients, and may obviate the need for jugular or subclavian catheterization in those who have undergone bidirectional cavopulmonary anastomosis.