Comparison of contrast and noncontrast magnetic resonance angiography for quantitative analysis of thoracic arteries in young patients with congenital heart defects

BACKGROUND: Contrast MRA (C-MRA) is the standard for quantitative analysis of thoracic vessels. We evaluated a noncontrast MRA (NC-MRA) sequence (3-D EKG and navigator-gated SSFP) for quantitative evaluation of the thoracic aorta and branch pulmonary arteries in young patients with congenital heart disease. OBJECTIVE: To compare contrast and noncontrast magnetic resonance angiography for quantitative analysis of thoracic arteries in young patients with congenital heart defects. METHODS: Measurements of thoracic aorta and branch pulmonary arteries were obtained from C-MRA and NC-MRA images in 51 patients, ages 2–35 years. Vessel diameters were compared using correlation and Bland-Altman analysis. Interobserver variability was assessed using percent variation. RESULTS: C-MRA and NC-MRA measurements were highly correlated (r = 0.91-0.98) except for the right pulmonary artery (r = 0.74, 0.78). Agreement of measurements was excellent (mean difference –0.07 to –0.53 mm; mean % difference –1.8 to –4.9%) except for the right pulmonary artery which was less good (mean difference 0.73, –1.38 mm; –3, –10%). Interobserver variability ranged from 5% to 8% for aortic and from 10% to 16% for pulmonary artery measures. The worse agreement and greater variability of the pulmonary artery measures appears due to difficulty standardizing the measurements in patients with abnormal and irregular vessels. CONCLUSION: These data indicate that C-MRA and NC-MRA measures are comparable and could be used interchangeably, avoiding administration of contrast in selected patients.