Objective: The aims of this study were to compare 3 contrast agents and to define feasible doses for quantitative lung perfusion imaging using the dual-bolus approach in dynamic contrast-enhanced (DCE) magnetic resonance imaging (MRI).
Materials and methods: Ten healthy volunteers (6 males, 4 females; mean age, 23.5 years) underwent DCE-MRI at 1.5 T using a 3D FLASH sequence. After a prebolus, 3 doses of gadopentetate dimeglumine (Gd-DTPA), gadofosveset, and gadobenate dimeglumine (Gd-BOPTA) were evaluated. Dose regimes were as follows: Gd-DTPA: 3.0 mL, 6.0 mL, and 12.0 mL with 1.5 mL prebolus; gadofosveset: 1.5 mL, 3.0 mL, and 6.0 mL with 0.8 mL prebolus; and Gd-BOPTA: 1.5 mL, 3.0 mL, and 6.0 mL with 0.8 mL prebolus. Pulmonary blood flow (PBF), pulmonary distribution volume, and mean transit time were assessed for each bolus. Region of interest measurements were used to determine the arterial input function (AIF) in the pulmonary trunk and signal intensities in lung parenchyma. Two radiologists independently rated the subjective image quality of the quantitative perfusion maps based on a 4-point Likert scale.
Results: Dose-dependent signal saturation effects were observed for all 3 contrast agents concerning AIF and parenchyma measurements. Signal yields were comparable using Gd-BOPTA (AIF, 214.49 arbitrary units [AU]; parenchyma, 41.7 AU) and Gd-DTPA (207.43 AU; 36.3 AU). Gadofosveset showed significantly lower signal yield (165.74 AU; 25.2 AU; p < 0.008). Highest signal increase was observed for Gd-DTPA. Using Gd-DTPA, mean PBF values for the 3 doses (3 mL, 6 mL, 12 mL) in mL/min per milliliter lung volume were 2.9 ± 1.5, 2.4 ± 1.1, and 1.6 ± 1.0. For the 3 doses of gadofosveset (1.5 mL, 3 mL, 6 mL) mean PBF results were 3.1 ± 1.1, 1.9 ± 0.7, and 1.2 ± 0.6. Last, mean PBF values for Gd-BOPTA (1.5 mL, 3 mL, 6 mL) were 3.4 ± 1.7, 2.8 ± 1.3, and 2.0 ± 0.8. Measurements provided consistent values for all perfusion parameters (PBF, pulmonary distribution volume, mean transit time) when compared with reference literature. Contrast dose volume and the applied contrast agent had no relevant effects on the image quality scores.
Conclusions: The dual-bolus approach using a 3D FLASH sequence is a feasible tool for quantitative lung perfusion imaging. Small boluses of 3 mL for Gd-DTPA, 1.5 mL for Gd-BOPTA, and 1.5 mL for gadofosveset provide sufficient signal yield for quantitative parenchyma measurements. Using higher boluses falsely lower perfusion values have to be considered due to signal saturation effects. Although gadofosveset yielded the lowest signal, the generated quantitative perfusion maps were of diagnostic quality.