To investigate the impact of bolus kinetics and diagnostic image quality of gadopentetate dimeglumine and gadobutrol in time-resolved thoracoabdominal 4D-MRA compared to dynamic computer tomographic angiography (dCTA).4-dimensional magnetic resonance angiography (4D-MRA) has been tested for a broad spectrum of indications and is increasingly used in clinical practice (1-5). It has been shown that physico-chemical properties including relaxivity, bolus application and concentration of gadolinium affect quantitative and qualitative parameters of MRA including vessel-to-background contrast (6-8). Compared to standard 0.5M gadopentetate dimeglumine, 1.0M gadobutrol provides a more compact bolus shape with significantly higher peak Gd concentrations leading to higher vessel signals and higher overall image quality (9-10). This study was performed to intra- and inter-individually analyze the impact of bolus kinetics including absolute Gd quantification (obtained by dCTA) on vessel-to-background contrast and image quality in 4D-MRA.7 anaesthetized Goettingen minipigs (46.1±4.0kg) underwent dCTA on a clinical dual source CT scanner (Definition, Siemens Healthcare, Erlangen/Germany) and 4D-MRA on a 3.0T whole-body scanner (Intera, Philips Healthcare, Best/the Netherlands). The animals were handled in compliance with the German animal welfare legislation and with the approval of the state animal welfare committee. All animals underwent dCTA and 4D-MRA using half (HGad; 0.05 mmol/kg b.w.) and standard doses (SGad; 0.1 mmol/kg b.w.) of 1.0M gadobutrol (Gadovist, Bayer Healthcare, Berlin, Germany) both administered at a rate of 1 mL/s. 4D-MRA sequences were additionally performed after standard doses (SMag; 0.1 mmol/kg b.w.) of 0.5M gadopentetate dimeglumine (Magnevist, Bayer Healthcare) injected with 2 mL/s. All CA applications were followed by 20 mL saline flushes at corresponding flow rates. 4D-MRA using 4D-TRAK was first performed with a highly temporally resolved, transversal image acquisition: TR, 7.7 ms; TE, 1.27 ms; FA, 25°; voxel size, [1.6×1.6×6.0] mm³; 200 dynamics, 16 slices each; 0.3s image update time; FOV, 410×410mm²; keyhole percentage, 25%; parallel imaging, SENSE, P(ap)=3, S(fh)=2; acquisition time (AQ), 60s. For SGad and HGad the arterial bolus passage in the descending aorta was intra-individually compared to dCTA acquisitions (80kV/150mAs, 0-40s, 0.3s/dynamic frame) performed at the same anatomical region. The CT signal enhancement was normalized to absolute Gd-concentration based on previous phantom experiments (2). Quantitative peak analysis included measurements of the arterial bolus peak height, length and full-width at half maximum (FWHM) in both dCTA and 4D-MRA. For qualitative image analysis, coronal isotropic 4D-MRA was separately acquired (voxel size, [1.5×1.5×1.5] mm³; 53 slices, 40 dynamics; 1.3 s image update time; AQ, 52s; all other scan parameters were identical to transverse 4D-MRA). Qualitative image analysis based on vessel visibility in 19 arterial and venous segments was performed by three readers. Comparing HGad to SGad, dCTA revealed a 39.0% lower bolus peak Gd concentration, 32.4% shorter bolus length and 20.6% shorter FWHM for HGad (Fig. 1A). Comparing HGad to SGad in highly temporally resolved transverse 4D-MRA, the peak was 14.5% lower, bolus length was reduced by 20.6% and FWHM shortened by 33.8%. In general, bolus curves were broader in 4D-MRA compared to dCTA. In 5/7 animals receiving HGad, 4D-MRA resulted in cut-off 1st pass arterial bolus peaks probably due to saturation effects (Fig. 1B). As a result, the arterial 1st pass peak signal was only 10% higher than the arterial 2nd pass peak. Bolus analysis of isotropic, coronal 4D-MRA revealed similar results. The arterial bolus peak signals were in the same range for all three CA applications (Fig.2). Venous bolus passage peak signals were highest after application of SGad (1.25x higher than SMag and 1.4x higher than HGad, Fig.2). Quality analysis (coronal 4D-MRA) showed an excellent inter-observer agreement for all CA applications (0.92-0.93). The qualitative analysis of vessel segments showed no significant difference in image quality between SGad and HGad (p=0.61). Analysis of vessel segment visibility after application of SMag was rated significantly lower compared to SGad and HGad (p<0.001). In addition, the analysis of venous vessel segments revealed a strong tendency for better vessel visibility after application of SGad compared to HGad (p=0.07).The peak Gd-concentration during the 1st arterial passage was higher after administration of standard-dose compared to half-dose gadobutrol. The respective peak signal intensity in 4D-MRA was comparable for both gadobutrol doses and standard dose gadopentetate dimeglumine due to peak cut-off effects at high vascular Gd concentrations. The venous phase of 4D-MRA revealed markedly higher peak signal intensities after administration of standard-dose gadobutrol, which is desirable in 4D-MRA, where all phases of contrast enhanced MRA are of interest. The overall image quality was rated significantly higher in 4D-MRA using gadobutrol compared to gadopentetate dimeglumine.