Muscle and brain spectroscopy

Hermien E. Kan, Beatrijs H.A. Wokke and Andrew G. Webb


MR spectroscopy of skeletal muscle provides a wealth of information on cellular metabolism. While 1H MRS is commonly used to assess levels of intramyocellular lipids in several diseases, 31P MRS can be used to study high energy phosphoryl metabolism and 13C MRS can be applied in the assessment of mitochondrial fluxes and dynamics of glycogen and creatine during resting conditions and exercise. At a field strength of 7T, not only the signal to noise is increased considerably compared to clinical field strengths (1.5 and 3T), but perhaps more importantly, also the spectral resolution is increased. At the C.J. Gorter Center, we aim to take advantage of these high field characteristics to study skeletal muscle metabolism in further detail


Creatine (Cr) and intra myocellular lipids (IMCL) are central compounds in energy metabolism in skeletal muscle. IMCL is an important metabolite in obesity and diabetes research while Cr, as it commonly shows little inter-individual and regional variation in normal volunteers during resting conditions, is often used for internal calibration to obtain absolute concentrations for other metabolites. In 1H MRS of skeletal muscle, the visibility of the signals of these two compounds is influenced by the angle of the muscle fibers with respect to the main magnetic field. Unfortunately, the optimal angle for these metabolites differs, making it difficult to reliably quantitate concentrations of both compounds in the same voxel. At the magic angle, for instance, in the soleus muscle (SOL), the signal from extra myocellular lipids (EMCL) partly overlaps the IMCL signal due to the shift of EMCL caused by bulk susceptibility effects. In contrast, the CH2 and CH3 resonances of Cr and phosphocreatine are easily visible at this angle, since the residual dipolar couplings that cause splitting of these signals is zero at this angle. A an angle of 0 degrees, however, in the tibialis anterior (TA) muscle, the separation between IMCL and EMCL is largest, but the Cr signal displays maximum dipolar splitting. Since the tri methyl ammonium (TMA) resonance commonly overlaps the upfield satellite peak of Cr at clinical field strengths (3 Tesla and below), quantification of Cr is challenging. By obtaining 1H MR spectra of the TA muscle at field strength of 7T the full triplet of the CH3 resonance of Cr is resolved, and IMCL and EMCL signals are separated due to the increased spectral resolution. This significantly increases the possibilities for accurate quantification of both compounds, using Cr as an internal reference. 1H MRS of the TA muscle is applied in studies of type II diabetes and healthy aging in collaboration with the department of Gerontology and Geriatrics.

Muscle Spectroscopy 1


31P MRS in the C.J. Gorter Center is mainly applied to study mitochondrial energy metabolism. To this end, we aim to develop and MR compatible exercise setup to study phosphocreatine recovery after in-magnet exercise. By performing this study at 7T, we can obtain signals from different muscles with high resolution.

Additionally, we aim to develop a new measure for non-invasive determination of mitochondrial content during resting conditions. 31P MRS approaches to obtain information on this parameter at clinical field strengths typically require in-magnet exercise thereby limiting application in a clinical setting. Due to the pH difference between the cytosol and the mitochondrial matrix, theoretically, two signals for inorganic phosphate (Pi) should be visible in the MR spectrum. If the concentration of Pi in the mitochondria scales with mitochondrial content, detection of this signal in resting muscle would constitute a much simpler alternative assessment of mitochondrial content.

Recently, we observed a peak 0.4 ppm downfield from the cytosolic Pi resonance (Pi1) in resting skeletal muscle using the increased signal-to noise and spectral resolution that can be obtained at 7 tesla (Kan H.E. et al, in press). This signal (which we term Pi2), at 5.1 ppm was attributed to the Pi pool inside the mitochondrial matrix based on various considerations including the chemical shift, the locations from which the signals were recorded, the short T1 relaxation time and differences between the tibialis anterior and the soleus muscle, muscles with different oxidative capacities. One of the aims of our 31P work is to further confirm the origin of the signal, as well as to optimize acquisition parameters for routine use.

Muscle Spectroscopy 2


Hermien Kan, PhD
C.J. Gorter Center for High Field MRI
Department of Radiology - C3Q
Leiden University Medical Center
Albinusdreef 2
2333 ZA Leiden 
The Netherlands
Phone: +31 71 5266097
Fax: +31 71 5248256