Dose-dependence of transient respiratory motion artifacts on gadoxetic acid-enhanced arterial phase MR images

Hiroyuki Morisaka1,2, Utaroh Motosugi1, Shintaro Ichikawa1

  • 1Department of Radiology, University of Yamanashi, Shimokato, Chuo-shi, Yamanashi, Japan.

Abstract

Insights

Higher doses of gadoxetic acid increase transient respiratory motion (TM) artifacts during arterial phase MRI. This study found a significant dose-dependence, with a nearly five-fold increased risk of moderate-to-severe artifacts at the higher dose.

Area of Science:

  • Radiology
  • Medical Imaging
  • Pharmacology

Background:

  • Transient respiratory motion (TM) artifacts can degrade image quality in dynamic MRI.
  • Gadoxetic acid is a contrast agent used in liver MRI.
  • Understanding artifact causes is crucial for accurate diagnosis.

Purpose of the Study:

  • To compare the incidence of TM artifacts between 0.05 mmol/kg and 0.025 mmol/kg gadoxetic acid doses.
  • To evaluate the dose-dependence of gadoxetic acid on TM artifact occurrence during arterial phase MRI.
  • To assess the intra-individual impact of different gadoxetic acid concentrations on artifact levels.

Main Methods:

  • Retrospective analysis of 91 patients undergoing dynamic 1.5T MRI.
  • Comparison of scans with 0.05 mmol/kg versus 0.025 mmol/kg gadoxetic acid.
  • Two radiologists graded TM artifacts on a four-point scale in arterial phase images.

Main Results:

  • Significantly more TM artifacts occurred with the 0.05 mmol/kg dose (17%) compared to the 0.025 mmol/kg dose (4% or 3%).
  • The odds ratio for moderate-to-severe artifacts was 4.99-5.33 with the higher gadoxetic acid dose.
  • A clear dose-dependence of gadoxetic acid on TM artifact occurrence was observed.

Conclusions:

  • Gadoxetic acid dose is directly related to the occurrence of transient respiratory motion artifacts.
  • Lower doses of gadoxetic acid may reduce TM artifacts in arterial phase MRI.
  • Optimizing contrast agent dosage can improve diagnostic accuracy in liver imaging.