Suppression of skeletal muscle signal using a crusher coil: A human cardiac 31p-MR spectroscopy study at 7 tesla

Abstract

Purpose The translation of sophisticated phosphorus MR spectroscopy (31P-MRS) protocols to 7 Tesla (T) is particularly challenged by the issue of radiofrequency (RF) heating. Legal limits on RF heating make it hard to reliably suppress signals from skeletal muscle that can contaminate human cardiac 31P spectra at 7T. We introduce the first surface-spoiling crusher coil for human cardiac 31P-MRS at 7T. Methods A planar crusher coil design was optimized with simulations and its performance was validated in phantoms. Crusher gradient pulses (100 μs) were then applied during human cardiac 31P-MRS at 7T. Results In a phantom, residual signals were 50 ± 10% with BISTRO (B1-insensitive train to obliterate signal), and 34 ± 8% with the crusher coil. In vivo, residual signals in skeletal muscle were 49 ± 4% using BISTRO, and 24 ± 5% using the crusher coil. Meanwhile, in the interventricular septum, spectral quality and metabolite quantification did not differ significantly between BISTRO (phosphocreatine/adenosine triphosphate [PCr/ATP] = 2.1 ± 0.4) and the crusher coil (PCr/ATP = 1.8 ± 0.4). However, the specific absorption rate (SAR) decreased from 96 ± 1% of the limit (BISTRO) to 16 ± 1% (crusher coil). Conclusion A crusher coil is an SAR-efficient alternative for selectively suppressing skeletal muscle during cardiac 31P-MRS at 7T. A crusher coil allows the use of sequence modules that would have been SAR-prohibitive, without compromising skeletal muscle suppression.