Jürgen Machann1, Malte Niklas Bongers2, Andreas Fritsche3, Hans-Ulrich Häring3, Mike Notohamiprodjo4, Andreas Greiser5, Konstantin Nikolaou4, and Fritz Schick2
1Section on Experimental Radiology, Department of Diagnostic and Interventional Radiology, Institute for Diabetes Research and Metabolic Diseases (IDM) of the Helmholtz Center Munich, German Center for Diabetes Research (DZD), Tübingen, Germany, 2Section on Experimental Radiology, Department of Diagnostic and Interventional Radiology, University Hospital Tübingen, Tübingen, Germany, 3Department of Endocrinology and Diabetology, Angiology, Nephrology and Clinical Chemistry, Institute for Diabetes Research and Metabolic Diseases (IDM) of the Helmholtz Center Munich, German Center for Diabetes Research (DZD), Tübingen, Germany, 4Department of Diagnostic and Interventional Radiology, University Hospital Tübingen, Tübingen, Germany, 5Siemens Healthcare, Erlangen, Germany
Synopsis
1H-MRS is increasingly applied
in many organs for non-invasive tissue characterization, e.g. for
quantification of ectopic lipids. Spectroscopic examinations of the myocardium
often suffer from limited spectral dispersion, thus limiting the metabolic information
content. Applying a new 60-channel body-array receive coil, high quality spectra
with superior dispersion as compared to previous setups are shown in this work.
A single voxel PRESS technique was applied in 10 subjects. After higher-order
shimming, linewidths of <20 Hz were obtained with high SNR in a clinically acceptable
measuring time. High reproducibility and performance of the method may promote 1H-MRS
applications in metabolic research and sports medicine.Purpose
1H-MRS of the myocardium is of
increasing interest with manifold applications e.g. in metabolic research [1-3]
or in sports medicine [4,5]. Acquisition of spectra is commonly performed
applying ECG triggering and respiratory gating in
order to avoid displacement of the volume of interest (VOI) and to obtain
sufficient SNR. Despite this double-triggered acquisition scheme, the quality
of spectra regarding spectral dispersion is typically rather limited – probably
due to inherent magnetic field inhomogeneities (coil characteristics).
Therefore signals of creatine (methyl-group, Cr
3 at 3.05 ppm) and
TMA (including choline and taurine at 3.2 ppm) often cannot be distinguished
and separation of different lipid signals is aggravated. Aim of this
preliminary study was to test the impact of a new 60-channel body-array receiver
coil on spectral resolution and SNR in cardiac
1H-MRS in comparison to an
18-channel body-array coil in combination with a spine-array coil.
Methods
10 healthy untrained male subjects (age: 22-48 years, BMI<30kg/m²)
were examined on a 3T whole body imager (Magnetom Skyra, Siemens Healthcare, Erlangen, Germany).
Volunteers were lying in supine position on the posterior part of the new
body-array coil and were covered by the anterior part (30 channels
each) and spine-array coil + 18 channel body-array coil, respectively, for
comparison. After morphological imaging including short-axis, four chamber-views
and cine imaging, a VOI of 7 ml (20x10x35mm³) was placed in the
intraventricular septum of the end-systolic cine images (see Figure 1a). Volume
selective shimming was performed by automated, volume-selective
higher-order
shimming for optimal homogeneity using a shim
volume of 4x4x4cm³. A PRESS-sequence with
following measurement parameters was applied: TE/TR 36/≈2000 ms, BW 2000 Hz, 8
acq. without water suppression (WS), 32 acq. with WS. TA between 30 s and 40 s
for spectra without WS and between 2:15 min and 2:40 min. for spectra with WS. Spectra
of the single channels were summed without further corrections. Post processing
included zero-filling to 4k, Fourier transformation and phase correction. Line
width of the water signal was determined and spectral dispersion of the
metabolites was rated by the resolution of Cr
3 and TMA. SNR was
calculated from raw data as ratio between the maximum signal of water divided
by the standard deviation of noise from unsuppressed spectra.
Results
A typical spectrum of a 25-year old male subject is shown in Figure 1b
(without WS) and 1c (enlarged, with WS). High SNR and
small linewidths enabled reliable separation of all
metabolites of interest in all subjects. A splitting of the two lipid
resonances – i.e. extramyocellular (EMCL) and intramyocellular lipids (IMCL) – can
be seen in some volunteers which enables differentiation and quantification of
these compartments. Line width of the water resonance ranged
between 13 Hz and 17 Hz in the subjects and all spectra, were of high quality with
excellent separation of Cr
3 and TMA (at least to 50% of the maximum
amplitude) and without any artifacts. SNR was clearly better (20-25%) and a
smaller line width of the water signal (1-3 Hz) was obtained for the new coil compared
to the 18 channel body-array coil in combination with the spine-array coil.
Discussion
Cardiac MRS can be seen as
the most challenging discipline for spectroscopic examinations in humans. A
stable position of the VOI during data acquisition despite heart beat and
breathing is a prerequisite for spectra of sufficient quality thus requiring
ECG triggering and respiratory gating. Furthermore, HF transmission and signal
reception – i.e. the applied coil – play a central role. The new 60-channel
body array coil provides essential advantages compared to other commercially
available coils with 16-32 channels and results in spectra with highest
quality. All metabolites – including EMCL and IMCL – can be quantified reliably
and offer a wealth of information about metabolism of the myocardium. It has to
be mentioned that IMCL and EMCL could not be separated in all volunteers
probably due to oblique muscle fiber orientation in the VOI under investigation
and/or due to a dominant EMCL or IMCL resonance. Cross-sectional and
interventional studies in the field of diabetes research and sports medicine
are planned for the future.
Conclusion
Spectra of the myocardium
could be assessed with excellent quality, enabling reliable quantification of
all metabolites of interest. Regarding spectral dispersion, the properties are
comparable to spectra recorded from the lower leg (e.g. soleus muscle),
resulting in improved possibilities regarding metabolic analyses for future
studies.
Acknowledgements
The study was supported in part by grants from the
DeutscheForschungsgemeinschaft (KFO 114), the German Federal Ministry of
Education and Research (BMBF) to the German Centre for Diabetes Research
(DZD). The authors thank Siemens Healthcare (Erlangen, Germany) for continuous support.References
[1] Ith M, Stettler C, Xu J, Boesch C, Kreis R. Cardiac lipid levels show diurnal changes and long-term variations in healthy human subjects. NMR Biomed. 2014;27(11):1285-1292.
[2] Krššák M, Winhofer Y, Göbl C et al. Insulin resistance is not associated
with myocardial steatosis in women. Diabetologia 2011;54(7):1871-1878.
[3] Wei J, Nelson MD, Szczepaniak EW et al. Myocardial Steatosis as a Possible Mechanistic Link
between Diastolic Dysfunction and Coronary Microvascular Dysfunction in Women. Am J Physiol Heart Circ Physiol
2015; doi: 10.1152/ajpheart.00612.2015
[4] Sai E, Shimada K, Yokoyama T et al. Association between myocardial triglyceride content and
cardiac function in healthy subjects and endurance athletes. PLoS One
2013;8(4):e61604
[5]
Bucher J, Krüsi M, Zueger T et al. The effect of a single 2 h bout of aerobic exercise on
ectopic lipids in skeletal muscle, liver and the myocardium.
Diabetologia
2014;57(5):1001-1005.