The findings from our study demonstrate that: 1) gadobutrol has significantly higher SI than gadoterate meglumine at the MAAo, AArch, and DAo; 2) CRs with respect to the trapezius muscle between gadobutrol and gadoterate meglumine were not significantly different at the MAAo, AArch, and DAo; 3) the maximal intraluminal orthogonal diameter measurements were not significantly different at different planes measured; 4) evaluation of overall image quality, wall conspicuity, and artifacts did not show any significant difference between the two GBCAs at the aortic root, MAAo, AArch, or the DAo.
Previous studies have compared the performance of gadobutrol and gadoterate meglumine for CE-MRA demonstrating variable results at other body regions (13–20). Wuesten et al. found that, for low-dose, high temporal resolution CE‐MRA of carotid artery aneurysms at 3T, gadobutrol was superior to gadoterate meglumine in terms of contrast-to-noise ratios (CNR) and depiction of morphological details (13). Similar observations were made by Morelli et al., who found that for renal artery stenosis assessment at 3T, low-dose gadobutrol (0.05 mmoL/kg) CE-MRA resulted in improved accuracy relative to equivalently dosed gadoterate meglumine (14). Kramer et al. found that, at equimolar doses, gadobutrol demonstrated higher signal-to-noise ratios (SNR), CNR, and superior image quality than gadoterate meglumine for dynamic and static carotid CE-MRA at 3T (15). Hansmann et al. found that, for equimolar, low-dose, time-resolved CE-MRA protocol of the calves, the significantly better SNR and CNR provided by gadobutrol compared to gadoterate meglumine did not translate into substantial differences in image quality (16). Using digital subtraction angiography (DSA) as the reference standard, Loewe et al. found that gadoterate meglumine was not inferior to gadobutrol in terms of diagnostic performance in patients with peripheral arterial occlusive disease undergoing CE-MRA at equimolar dose at 3T (17). Haneder et al. found that, at equimolar doses, gadobutrol yielded significantly higher SNR and CNR while gadoterate was better rated in terms of overall image quality and diagnostic confidence for the evaluation of peripheral arterial occlusive disease (18). Hoelter et al. found that, gadobutrol resulted in a significantly higher SNR/CNR and better delineation of the intracranial vasculature when compared to gadoterate meglumine for cervical and cerebral CE-MRA at 1.5T (19). Lee et al. found that, at equimolar doses, increased gadolinium delivery over time using gadobutrol provides higher relative enhancement parameters in benign prostatic hyperplasia nodules compared with gadoterate meglumine but does not translate into improved SNR or CNR (20). However, results from CE-MRA studies of certain anatomical areas are not transferable to other areas without careful interrogation because the injection volume, injection rate, and the contrast agent travel time influence the concentration of GBCA in the region of interest significantly.
Due to their paramagnetic nature, GBCAs shorten the tissue relaxation time, resulting in increased tissue signal intensity on T1-weighted images (23). A higher relaxivity would amplify the T1 shortening effects, enabling CE-MRA to be performed at lower doses with a potential reduction of the risk of GBCA-induced toxicity. Previously, numerous studies in adult patients have shown that gadobutrol, due to its relatively higher r1 relaxivity (5.3 L mmol− 1s− 1 in blood at 37°C/1.5 T) (24), results in significantly improved image quality and diagnostic performance relative to that achieved with comparator GBCAs at equivalent dose. When compared to gadobutrol, gadoterate meglumine has a lower r1 relaxivity (4.2 L mmol− 1s− 1 in blood at 37°C/1.5 T) (24). The higher SI of gadobutrol when compared to gadoterate meglumine as demonstrated in the current study is likely due to these differences in relaxivities. This may also be more marked in the thoracic aorta as the contrast bolus is not significantly dispersed since it is close to the heart, unlike other more peripheral vascular regions, where the contrast bolus is more dispersed. However, the CR of the intravascular compartment with respect to the trapezius muscle showed no significant difference between the contrast agents in our study. This may reflect that part of the difference in the SI between the two exams maybe related to technical and/or patient-related differences as the two exams were done on separate days and separate machines. Another factor to consider for first-pass CE-MRA is edge blurring. Theoretically, edge blurring occurs if only the center parts of k-space are sampled during the presence of contrast agent in the region of interest and could be enhanced while using gadobutrol due to the smaller injected volume when compared to gadoterate meglumine. However, this was neither quantitatively measured nor reflected as impaired qualitative scores in our study. This suggests that 0.2 mmol/kg bodyweight of a 1.0 molar GBCA injected at 2.0 mL/sec has a sufficiently long contrast agent bolus to avoid disturbing edge blurring, i.e., contrast agent is present in the vascular territory of interest during the entire k-space sampling to an adequate degree.
In our study, a double dose of GBCAs was used for intraluminal contrast enhancement. This was because the standard-of-care clinical MRA was always performed in conjunction with a comprehensive cardiac MRI protocol. For research MRA, we matched the GBCA dose to the clinical MRA. Previous studies have demonstrated the safety (25) and superior signal-to-noise (26, 27) on higher dosing of GBCAs, but its effects on clinical practice may be negligible. Furthermore, in our study, we matched the flow rate of the two GBCA (2.0 mL/s). This may have led to a higher peak concentration of GB when compared to GM, influencing the SI measurements. Future studies matching the gadolinium injection rates have to be performed to evaluate the impact of flow rates on SI and image quality.
This study had some limitations. First, the size of our study population is relatively small. Prospective patient acquisition was difficult as they had to fulfill inclusion criteria as mentioned above. Second, we were not able to randomize patients to the order in which they received the two different GBCAs, as patient enrollment occurred after their standard-of-care gadobutrol-enhanced MRA. Our institution uses gadobutrol for clinical CE-MRA of the thoracic aorta. Third, although all the examinations were performed on 1.5-Tesla MR scanners, for some subjects, the clinical CE-MRA were performed on a different model (Avanto or Aera) than the research CE-MRA (always Aera).
In conclusion, this prospective intraindividual study comparing equal doses of gadobutrol and gadoterate meglumine demonstrates that for CE-MRA of the thoracic aorta, the CR with respect to muscle, aortic luminal diameter measurements, and qualitative assessment including overall image quality, vessel-wall conspicuity, and artifacts are comparable between the two GBCA. However, gadobutrol demonstrated a significantly higher aortic luminal SI when compared to gadobuterol meglumine. Future studies are warranted to evaluate the clinical impact of improved SI of gadobutrol for the assessment of the thoracic aorta.