19F MRI Imaging Strategies to Reduce Isoflurane Artifacts in In Vivo Images

• Published in: Molecular imaging and biology Vol. 24; no. 1; pp. 71 - 81
• Main Authors: Staal, Alexander H. J., Veltien, Andor, Srinivas, Mangala, Scheenen, Tom W. J.
• Format: Journal Article
• Language: English
• Published: Cham Springer International Publishing 01.02.2022; Springer Nature B.V
• Subjects: Bandwidths; Brief; Brief Article; Chemical equilibrium; Contrast agents; Contrast media; Excitation; Fluorine; Image contrast; Image processing; Imaging; In vivo methods and tests; Inhalation; Isoflurane; Magnetic resonance imaging; Medicine; Medicine & Public Health; Radiology; Respiration; Signal to noise ratio; Artifacts; Anesthesia; Chemical shift; Imaging; Isoflurane; Fluorine-19 Magnetic Resonance Imaging
• ISSN: 1536-1632; 1860-2002
• DOI: 10.1007/s11307-021-01653-6

Purpose Isoflurane (ISO) is the most commonly used preclinical inhalation anesthetic. This is a problem in 19 F MRI of fluorine contrast agents, as ISO signals cause artifacts that interfere with unambiguous image interpretation and quantification; the two most attractive properties of heteronuclear MRI. We aimed to avoid these artifacts using MRI strategies that can be applied by any pre-clinical researcher. Procedures Three strategies to avoid ISO chemical shift displacement artifacts (CSDA) in 19 F MRI are described and demonstrated with measurements of 19 F-containing agents in phantoms and in vivo ( n  = 3 for all strategies). The success of these strategies is compared to a standard Rapid Acquisition with Relaxation Enhancement (RARE) sequence, with phantom and in vivo validation. ISO artifacts can successfully be avoided by (1) shifting them outside the region of interest using a narrow signal acquisition bandwidth, (2) suppression of ISO by planning a frequency-selective suppression pulse before signal acquisition or by (3) preventing ISO excitation with a 3D sequence with a narrow excitation bandwidth. Results All three strategies result in complete ISO signal avoidance ( p  < 0.0001 for all methods). Using a narrow acquisition bandwidth can result in loss of signal to noise ratio and distortion of the image, and a frequency-selective suppression pulse can be incomplete when B 1 -inhomogeneities are present. Preventing ISO excitation with a narrow excitation pulse in a 3D sequence yields the most robust results (relative SNR 151 ± 28% compared to 2D multislice methods, p  = 0.006). Conclusion We optimized three easily implementable methods to avoid ISO signal artifacts and validated their performance in phantoms and in vivo . We make recommendation on the parameters that pre-clinical studies should report in their method section to make the used approach insightful.