Calculation of a subject-specific adaptive motion-correction factor for improved real-time navigator echo-gated magnetic resonance coronary angiography.

There has been conflicting data in the literature regarding the use of wide navigator echo (NE) acceptance windows in combination with adaptive motion correction for magnetic resonance coronary angiography (MRCA). This in part may be due to the use of a fixed correction factor when applying the adaptive motion-correction algorithm, which may potentially result in miscorrection of the imaging slice in subjects whose correction factor differs widely from the mean. We have addressed this issue by measuring the superior/inferior correction factor in 25 subjects and assessing the effect of using a subject-specific correction factor (CFss) for MRCA in comparison with no adaptive motion correction (CF0) and erroneous adaptive motion correction with a correction factor of 1.0 (CF1). There was a wide variation in the correction factor between subjects (proximal right coronary artery, 0.49 +/- 0.15, range 0.20-0.70; proximal left coronary artery, mean 0.59 +/- 0.15, range 0.20-0.85). The subject-specific correction factor was accurately calculated from motion of the aortic root in the coronal plane between expiratory and inspiratory breathhold (correction factor calculated from coronal image versus correction factor calculated after localization of coronary arteries, r = 0.92, p < 0.001). MRCA image quality was improved using a subject-specific correction factor, for both a 6-mm NE acceptance window (CFss versus CF0, p = 0.008; CFss versus CF1, p = 0.02) and a 16-mm NE window (CFss versus CF0, p = 0.01; CFss versus CF1, p = 0.007). Furthermore, image quality was maintained between the two NE windows if the subjects-specific correction factor was used (6 versus 16 mm, p = 0.21), with an improvement in scan efficiency (6 versus 16 mm, 49 +/- 17% versus 81 +/- 22% respectively, p < 0.001). Thus, for adaptive motion correction to be implemented, a subject-specific correction factor should be used and calculated from simple coronal expiratory and inspiratory breathholds. For real-time NE-gated cardiac MR with adaptive motion correction, the NE window can be widened to reduce the acquisition period without loss of image quality.