Magnetic Resonance Imaging MRI and spectroscopy

Positioning of measured samples in strong homogeneous magnetic fields forces protons within the sample to align according to the direction of the magnetic field and thus create a macroscopic magnetisation of the sample. Interaction of spins with externally applied radiofrequency (RF) waves at the resonant frequency changes the magnitude and direction

Visceral Adiposity Abdomen

Figure 13.3 Assessment of Abdominal Fat Storage by Computed Tomography (CT) Representative cross-sectional abdominal CT scans of a lean (A) and an obese (B) research volunteer, demonstrating the fat/muscle CT contrast shown with demarcations of visceral (large arrowheads), deep subcutaneous (open arrows) and superficial subcutaneous (closed arrows) adipose tissue (AT) depots. The fascia (small arrowhead) within subcutaneous abdominal AT was used to distinguish superficial from deep depot. In the two CT scans shown, the area of superficial subcutaneous AT was similar (144 vs 141 cm2), whereas areas of deep subcutaneous (126 vs 273 cm2) and visceral (84 vs 153 cm2) AT were quite different. Insulin-stimulated glucose metabolism was 6.1 and 4.0 mg/min-1/kg FFM-1 in lean and obese volunteers, respectively (FFM: fat-free mass). Reproduced from Kelley D E et al. (2000) Am J Physiol Endocrinol Metab 278 (5) E941-E948. Courtesy of the American Physiological Society.

Figure 13.3 Assessment of Abdominal Fat Storage by Computed Tomography (CT) Representative cross-sectional abdominal CT scans of a lean (A) and an obese (B) research volunteer, demonstrating the fat/muscle CT contrast shown with demarcations of visceral (large arrowheads), deep subcutaneous (open arrows) and superficial subcutaneous (closed arrows) adipose tissue (AT) depots. The fascia (small arrowhead) within subcutaneous abdominal AT was used to distinguish superficial from deep depot. In the two CT scans shown, the area of superficial subcutaneous AT was similar (144 vs 141 cm2), whereas areas of deep subcutaneous (126 vs 273 cm2) and visceral (84 vs 153 cm2) AT were quite different. Insulin-stimulated glucose metabolism was 6.1 and 4.0 mg/min-1/kg FFM-1 in lean and obese volunteers, respectively (FFM: fat-free mass). Reproduced from Kelley D E et al. (2000) Am J Physiol Endocrinol Metab 278 (5) E941-E948. Courtesy of the American Physiological Society.

of this macroscopic magnetisation, in a process called excitation. After switching off the RF waves, magnetisation returns to the thermal equilibrium in a time and direction dependent manner. This process is called relaxation and the changing macroscopic magnetisation induces an electrical signal in the receiver coil. Signal intensity in MRI is dependent not only on the proton density in the volume of interest, but also on the interaction between tissue protons and externally applied RF waves during the excitation, and on the interaction between nuclei within the tissue during the relaxation. The different physical properties of protons within water and fatty acid molecules result in strong differences in the time and phase dependent behaviours of these nuclei during the relaxation, producing a relaxation and phase dependent contrast for magnetic resonance imaging.

Computed Tomography Fatty Liver

Figure 13.4 Assessment of Liver Fat Accumulation by CT A) CT scan of normal liver. The liver is denser (brighter) than the spleen; B) CT scan of a fatty liver. The liver is less dense (darker) than the spleen. Reproduced with permission from Joy D et al. (2003) Eur J Gastroenterol Hepatol 15 (5), 539-543.

Figure 13.4 Assessment of Liver Fat Accumulation by CT A) CT scan of normal liver. The liver is denser (brighter) than the spleen; B) CT scan of a fatty liver. The liver is less dense (darker) than the spleen. Reproduced with permission from Joy D et al. (2003) Eur J Gastroenterol Hepatol 15 (5), 539-543.

T -relaxation based contrast can be used for volumetric measurements of subcutaneous and intra-abdominal fat accumulation. Multi-segment multi-slice 2D images or real 3D data sets are acquired (Thomas et al. 1998) and areas of fat volume can be segmented and divided into separate compartments in more or less automated fashion (Thomas & Bell 2003; Positano et al. 2004) (Figure 13.5). The acquisition of multi-segment multi-slice 2D data sets once required about 30 minutes, but the development of MRI hardware, data acquisition and reconstruction now allow for continuous whole body data acquisition during the defined movement of the patient table in the magnet (Kruger et al. 2002, 2005; Aldefeld et al. 2006; Sommer et al. 2006) (Figure 13.6). This improvement can reduce scanning time to less than three minutes and will make whole-body MRI measurement of fat distribution highly desirable and affordable, even for clinical praxis.

Phase-behaviour based contrast can be used for the quantitation of intrahepatic fat accumulation. Images with water and fat signal contributions 'in phase' and 'out of phase' are added or subtracted in order to obtain pure water and pure fat images (Dixon 1984; Reeder et al. 2004). Recently an iterative decomposition method yielding water and fat images from a single image acquisition was presented (Reeder et al. 2005; Fuller et al. 2006) and its feasibility for dynamic imaging was proven (Yu et al. 2006). Spatial resolution of MR images is given by system hardware performance and can be lower than 1 mm in plane.

In addition to imaging techniques, magnetic resonance can be applied in a spectro-scopic fashion. Different electron clouds within the molecule result in a different resonance frequency of protons in water and fatty acid. This effect is called chemical shift of resonant frequencies and it does not depend on the magnetic field strength applied. In praxis it is given in relative units - parts per million (ppm). Localisation of the spectroscopic signal can be achieved in single voxel (volume pixel) (Figure 13.7) or in matrix-based multi voxel fashion. In the latter case, the method is called chemical shift imaging (CSI) or spectroscopic imaging (SI). The signal intensities of different chemical entities can be used to produce distribution maps of metabolites of interest. Current best achievable geometric resolution is ~1 cm3 for the *H single voxel spectroscopy (Anderwald et al. 2002) and ~0.5 cm3 for spectroscopic imaging (Hwang et al. 2001; Vermathen et al. 2004) (Figure 13.8). The main

Slices Abdomen Mri

Figure 13.5 Assessment and Quantitation of Subcutaneous and Visceral Fat Amount by Multi-Slice MRI of Abdomen Selected representative MRI images acquired from the lower abdomen (lower left corner) up to the level of the liver (upper right corner) of a young overweight patient depict the T1 weighted MRI fat/muscle contrast. Middle slice in the belly region is enlarged (left) and the semiautomatic segmentation of different fat contributions (right; subcutaneous layer: grey; visceral fat: white) preceding the quantitation of fat volume is illustrated. Image data were kindly provided by M. Chmelfk, Medical University of Vienna.

Figure 13.5 Assessment and Quantitation of Subcutaneous and Visceral Fat Amount by Multi-Slice MRI of Abdomen Selected representative MRI images acquired from the lower abdomen (lower left corner) up to the level of the liver (upper right corner) of a young overweight patient depict the T1 weighted MRI fat/muscle contrast. Middle slice in the belly region is enlarged (left) and the semiautomatic segmentation of different fat contributions (right; subcutaneous layer: grey; visceral fat: white) preceding the quantitation of fat volume is illustrated. Image data were kindly provided by M. Chmelfk, Medical University of Vienna.

advantage of spectroscopic measurement is the direct separation and quantitation of water and fat fractions of diverse soft tissues. 13C MRS, even though not so sensitive regarding the spectroscopic volume of interest, can also easily distinguish saturated and unsaturated tissue fat contributions (Beckmann et al. 1992; Petersen et al. 1996; Thomas et al. 1997). Absolute fat quantitation by MRS has been validated against gold standard histological and biochemical measures or CT measurements for liver (Longo et al. 1993; Thomsen et al. 1994; Petersen et al. 1996; Szczepaniak et al. 1999) and myocardial lipid measurements (Szczepaniak et al. 2003).

Differences in the magnetic properties of bulk cylinders like extramyocellular lipids (EMCL) and spherical vesicle-accumulated intramyocellular lipids (IMCL) give the potential to separate these two compartments. This phenomenon was observed for the first time in the early 1990s (Schicket al. 1993) and confirmed by theory (Boesch et al. 1997; Boesch & Kreis 2001) and model experiments (Szczepaniak et al. 2002) later on (Figure 13.7). Successful validation by histological and biochemical analysis of biopsies (Howald et al. 2002) and the increasing accessibility of MR equipment have led to numerous metabolic studies ever since (Krssak & Roden 2004; Boesch et al. 2006). For quantification purposes, the signal of tissue water (Boesch et al. 1997; Szczepaniak et al. 1999) or skeletal muscle creatine (Rico-Sanz et al. 1999) is taken as the internal reference. Lipid concentration can be given in these relative units (Krssak et al. 1999) or, assuming certain hydration or creatine concentration in

Body Fat Distribution Mri

Figure 13.6 Whole Body Fat Distribution by Continuous Moving Table MRI Head-to-toe images acquired in two 3D MRI scans, with water-selective (a) and fat-selective (b) presaturation pulses. Each image shows one coronal plane from a 3D data set. The total scan time was 6 min per run. Reproduced with permission from Aldefeld B et al. (2006) Magn Reson Med 55 (5), 1210a-1216.

Figure 13.6 Whole Body Fat Distribution by Continuous Moving Table MRI Head-to-toe images acquired in two 3D MRI scans, with water-selective (a) and fat-selective (b) presaturation pulses. Each image shows one coronal plane from a 3D data set. The total scan time was 6 min per run. Reproduced with permission from Aldefeld B et al. (2006) Magn Reson Med 55 (5), 1210a-1216.

the tissue, fatty acid chain length and saturation distribution, and the specific weight of the water and fat compartment, calculated into normal concentration units (mmol/l-1; mg/kg-1) (Szczepaniak et al. 1999).

Another MR technique using conventional imaging technology, but exciting only the fat fraction of the tissue - based on its specific resonance frequency - is the recently introduced fat-selective MRI (Schick et al. 2002) (Figure 13.7). This method can produce fat distribution maps of liver and muscle with excellent spatial resolution (< 1 mm in plane)

Extramyocellular Lipid

a ppm b PPm

Figure 13.7 Comparison of Fat-Selective Imaging and Volume Localised Spectroscopy of Lipids Upper left panel shows a traditional T1 weighted MRI image, which guides the localisation of spectrum acquisition. Two volume elements of 12 mm X 12 mm X 20 mm were examined by a single-voxel STEAM technique. The spectra were recorded from representative regions of the tibialis anterior muscle (ROI I) and the soleus muscle (ROI II). The lipid signals in the spectra show a composition of EMCL and IMCL components. The water signal in spectra without water suppression (not shown) serves as reference for the assessment of total fat content. Upper right panel: analysis of the muscular lipid content in fat-selective images with 10 mm thickness was performed by the mean SI in selected ROIs (ROI I in the tibialis anterior muscle and ROI II in the soleus muscle). The borders of the ROIs can interactively be chosen during data processing in order to avoid undesired contributions from fatty material in the septa. Tibial bone marrow (Ref) serves as an internal reference with nearly 100% fat content. Reproduced with permission from Schick F et al. (2005) Magn Reson Med 47 (4), 720-727.

(Machann et al. 2003; Goodpaster et al. 2004) but cannot separate extramyocellular and intramyocellular signal contributions. A similar disadvantage is found in CT quantification of muscle fat by overall signal attenuation. A recent study found that CT measurement was correlated with total intramuscular lipid concentration measured by MR spectroscopy (EMCL & IMCL) (Larson-Meyer et al. 2006b). Splitting the MRS measure into extra- and intra-cellular compartments yielded greater CT prediction of EMCL in the tibialis anterior muscle and of IMCL in the soleus muscle (Larson-Meyer et al. 2006b). This observation can be explained by the prevailing amount of EMCL and IMCL in the respective muscle group (Jacob et al. 1999; Anderwald et al. 2002; Kautzky-Willer et al. 2003).

Muscle Proton Spectroscopy Mri

Figure 13.8 Assesment of Intramyocellular Fat Distribution by Proton Chemical Shift Imaging (CSI) Superb localisation (~0.33 cm3) of spectral information and resulting excellent spectral resolution (separation between EMCL and IMCL contribution) achieved by JH CSI at higher field strength (3T). JH MR spectra are simultaneously obtained from the volumes of interest in soleus muscle as depicted on the image. Particular spectra can be added, yielding higher signal-to-noise ratio and anatomically matched voxel distribution. Image data were kindly provided by S. Gruber & M. Chmelik, Medical University of Vienna.

Figure 13.8 Assesment of Intramyocellular Fat Distribution by Proton Chemical Shift Imaging (CSI) Superb localisation (~0.33 cm3) of spectral information and resulting excellent spectral resolution (separation between EMCL and IMCL contribution) achieved by JH CSI at higher field strength (3T). JH MR spectra are simultaneously obtained from the volumes of interest in soleus muscle as depicted on the image. Particular spectra can be added, yielding higher signal-to-noise ratio and anatomically matched voxel distribution. Image data were kindly provided by S. Gruber & M. Chmelik, Medical University of Vienna.

From a safety point of view, magnetic resonance techniques represent no radiation risk, but as discussed in Chapter 11, the presence of a strong magnetic field and the switching of magnetic field gradients make metallic objects (splinters, tattoos, coloured contact lenses, piercings, uterus coils), other medical devices (pace makers, cardiac valves, clips, electrodes, neuro-stimulators), implants, prosthetics, shunts and stents contraindication for the MR examination. Another practical consideration is the restricted space in the clear bore of the magnet. The usual clear diameter of ~60-70 cm can exclude morbidly obese patients from the examination. Nevertheless, the advantages and the versatility of the method as well as the wider spread of clinical MR systems predetermine broad application in future clinical praxis.

Supplements For Diabetics

Supplements For Diabetics

All you need is a proper diet of fresh fruits and vegetables and get plenty of exercise and you'll be fine. Ever heard those words from your doctor? If that's all heshe recommends then you're missing out an important ingredient for health that he's not telling you. Fact is that you can adhere to the strictest diet, watch everything you eat and get the exercise of amarathon runner and still come down with diabetic complications. Diet, exercise and standard drug treatments simply aren't enough to help keep your diabetes under control.

Get My Free Ebook


Post a comment