The purpose of this study was to develop a standardized method of 2D MRA of the proximal 50 mm of the major epicardial coronary arteries. Therefore the efficacy of fixed imaging planes (transverse, coronal, sagittal, 30-degree RAO equivalent, and 60-degree LAO equivalent) in imaging coronary arteries was compared to that of oblique planes defined by the operator on a previously obtained image. 3D data sets obtained by a respiratory-gated 3D MRA method in eight patients with a mean age of 57 years were studied by multiplanar reformatting. Efficacy of planes was expressed as an IOE. Fixed transverse imaging planes proved to be equally efficacious as operator-defined planes in imaging the left main (IOE 2.2 +/- 1.0 vs 2.2 +/- 0.9, p = NS) and LAD (IOE 6.0 +/- 1.9 vs 8.2 +/- 2.0, p = NS). Operator-defined planes were superior to fixed imaging planes in imaging the RCA (IOE 6.3 +/- 1.2 vs 3.5 +/- 1.2, p < 0.001) and the LCx (IOE 6.2 +/- 2.3 vs 4.8 < 2.3, p < 0.05). On the basis of these results, a standardized 2D MRA protocol for the proximal coronary arteries was proposed. Pitfalls in interpretation of coronary MRA images were discussed.