OBJECTIVE The aim of this study was to optimize contrast media dose for assessment of pulmonary circulation with dynamic time-resolved contrast-enhanced 3D MR angiography. SUBJECTS AND METHODS. Twenty healthy volunteers (20-38 years old; mean [+/- SD], 27.2 +/- 4.5 years) were examined prospectively using turbo fast low-angle shot MR angiography (TR/TE, 2.4/1.04). Ten consecutive coronal 3D slabs with a frame rate of 3.2-sec duration were acquired during injection of contrast media at a rate of 4 mL/sec. Signal intensities were measured in various vessels and pulmonary parenchyma. Maximum signal-intensity enhancement (DeltaSI(max)) and time to peak enhancement were calculated. Depiction of pulmonary vessels and pulmonary parenchyma was scored according to an image quality score. RESULTS Central pulmonary arteries were well visualized at all tested doses. Segmental arteries, however, were blurry with 0.025 or 0.05 mmol/kg; image quality was improved at 0.1 mmol/kg of gadoterate meglumine (p < 0.05). Image quality did not further improve at 0.2 mmol/kg (p = not significant). Values for DeltaSI(max) in the pulmonary trunk were 38.9 +/- 9.7, 64.1 +/- 9.1, 79.7 +/- 12.2, and 96 +/- 6.0 at 0.025, 0.5, 0.1, and 0.2 mmol/kg of gadoterate meglumine, respectively. Pulmonary parenchyma showed almost no enhancement at 0.025 and 0.5 mmol/kg of gadoterate meglumine (DeltaSI(max) = 1.6 +/- 1.1 and 1.6 +/- 1.2, respectively), but better visualization was shown with 0.1 and 0.2 mmol/kg of gadoterate meglumine (DeltaSI(max) = 2.9 +/- 0.8 and 6.7 +/- 2.1, respectively). Time from peak enhancement in pulmonary arteries to peak enhancement in veins was independent of dose. CONCLUSIONS A dose of 0.1 mmol/kg of gadolinium chelate allows depiction of pulmonary arteries and qualitative assessment of pulmonary parenchyma. Thus, 0.1 mmol/kg can be recommended for dynamic contrast-enhanced 3D MR angiography.