Respiratory Motion Artifacts During Arterial Phase Imaging With Gadoxetic Acid: Can the Injection Protocol Minimize This Drawback?

Purpose: To determine which of three gadoxetic acid injection techniques best reduced the contrast-related arterial-phase motion artifacts. Materials and Methods: This Institutional Review Board (IRB)-approved, retrospective study included a cohort of 78 consecutive patients who each had serial gadoxetic acid-enhanced 3.0T magnetic resonance imaging (MRI) of the liver (0.025 mmol/kg body weight) performed with at least two of three injection techniques: M1 test bolus, undiluted, power-injected 1mL/s; M2 test bolus, diluted 50% with saline, power-injected 1mL/s; M3 fixed delay, undiluted, manually injected. Blinded to the injection method, three readers independently rated the randomized images for arterial-phase motion artifacts, arterial-phase timing, and arterial-phase lesion visibility using a four-point Likert scale. Results: Regarding respiratory artifacts, gadoxetic acid arterial-phase images were judged better with M3 (2.7 +/- 0.7) and were significantly less than those with M1 (2.1 +/- 1.1) (P = 0.0001). Arterial-phase M2 (2.50 +/- 0.89) images were rated significantly better than arterial-phase M1 images (P = 0.012), but the difference between arterial-phase images with M3 and M2 scores was not statistically significant (P = 0.49). Arterial-phase timing was significantly better for M1 compared to M3, and for M2 compared to M3 (P < 0.0001 for both). The area under the curve was 0.59-0.68. However, there was no significant difference between M1 and M2 (P = 0.35). With regard to arterial-phase lesion visibility, there was no significant difference in the ratings between any of the three injection techniques (P = 0.29-0.72). Interreader agreement was moderate to substantial (kappa = 0.41-0.62). Conclusion: A diluted, power-injected protocol (M2) seems to provide good timing and minimize artifacts compared with two other injection methods. No significant difference was found in lesion visibility between these three methods.