Magnetic Resonance Spectroscopy (MRS) PDF

# Magnetic Resonance Spectroscopy (MRS): Principle and Applications ## By: Dr. Amal Alorainy ## Objective - Understand the Principles of Magnetic Resonance Spectroscopy (MRS). - Analyze Spectral Data and Chemical Shifts. - Explore Clinical Applications and Techniques of MRS. ## Introduction - F...

# Magnetic Resonance Spectroscopy (MRS): Principle and Applications ## By: Dr. Amal Alorainy ## Objective - Understand the Principles of Magnetic Resonance Spectroscopy (MRS). - Analyze Spectral Data and Chemical Shifts. - Explore Clinical Applications and Techniques of MRS. ## Introduction - Following the excitation & during relaxation, radiofrequency signals are generated & expressed as a frequency spectrum. - MRS is an analytical method using MRI to identify & quantify metabolite concentration in biological tissue. - It is different from conventional MRI in producing *Spectra* that provides physiological & chemical information instead of anatomical detail. ## Magnetic Resonance Spectroscopy - Used as a predictor of disease processes & treatment response. - Can obtain spectra from different *MR active nuclei*: - 1H protons used due to its high sensitivity & *MRI abundance*. - MR Spectrum displays the chemical composition of tissue by measuring the position of molecular peaks in the spectrum. ## Spectral Characteristics - Peak position along the NMR spectrum is *constant* and depends on its *chemical shift*. - The size and shape of a peak is variable and depends on five major factors: 1. the concentration of active nuclei. 2. the T1 and T2 relaxation times of the metabolite, also affected by the TR and TE of the MRS sequence. 3. magnetic inhomogeneity (T2*) effects across the ROI, controlled by shimming. 4. The presence of hidden/overlapping peaks. 5. Whether the peak is single or split into multiplets by a J-coupling interaction. If a peak is split into multiplets, the overall complex of sub-peaks will be smaller and broader than the original peak would have been without coupling. ## Why Are Some Spectral Peaks Taller Than Others? Why Are Some Wide While Others Are Narrow? - **TR controls T1 W**: When TR is much shorter than the T1 time of a metabolite, the longitudinal magnetization of that metabolite does not fully recover & its peak height will be reduced. - **TE controls T2 W**: T2 determines the rate of decay of the signal & the width of a spectral peak. Metabolites with shorter T2's will decay faster and have smaller peaks & wider widths than those with longer T2's. ## Chemical Shift - MRS uses signal intensity, line width (FWHM) & position to display chemical shift or resonant frequency difference between metabolites. - *Chemical shift* refers to the slight difference in resonance frequencies between two identical nuclei (1H) due to differences in their local magnetic environments (*chemical environments*) & is expressed in parts per million (ppm). - This frequency difference arises because the resonance frequency of a particular nucleus is not determined by the strength of the Bo, but by the local field experienced by the nucleus at the atomic level. - All nuclei therefore do not resonate at precisely the same frequency. For 1H nuclei chemical shifts are relatively small, but are nevertheless detectable. - Chemical shifts can provide important info about the molecular composition of the tissue imaged and form the basis of MRS. ## MR Spectrum - MRS: Plot of signal intensity (Y-axis) vs frequency (X-axis). - Area of each peak represents the relative number of protons in that particular position. - X-axis: Chemical shift of metabolites in ppm. Increases from right to left. ## Equipment Requirements - Frequency separation of spectral peaks depends on field strength (Bo) & homogeneity: - MRS requires high magnetic field strength, 1.5 T or more, and multi-element phased array RF coil. *WHY?* Because metabolites produce relatively weak signal, so we need to increase the SNR. - Shimming technique is essential to optimise magnetic field homogeneity over the voxel & improves water suppression. - Produces a readable spectrum. - Improves the sensitivity & resolution of metabolite signal by narrowing the peak widths & increasing SNR. ## Difference Between Good & Poor Shimming - **Optimised shimming:** Reduced line widths with proper shimming. - **Distorted shimming:** Broadened line widths. ## Signal Suppression - MRS requires suppression of signals from water & fat. - **Water Suppression:** - Necessary to avoid superimposing high water peak over the peaks from the lower-concentration brain metabolites, allowing them to be depicted on the spectrum. - Achieved using a very narrow BW Chemical Shift Selective Saturation (CHESS) pre-pulse. Before the MRS acquisition, narrow frequency-selective 90 RF pulses are applied at exactly the Larmor frequency of water, followed by dephasing gradient pulses to spoil any transverse magnetization. Suppress water signal. - **Fat Suppression:** - Sequence incorporating inversion pulses. ## Characteristics of MRS at 3T - Greater frequency separation between peaks (if peaks overlap) difficult to interpret & measure peak heights. - Higher spatial resolution increases distance & easier distinction between metabolite peaks. - Change in relaxation times at 3T influences metabolite signal appearance: - T1 relaxation times lengthen, this affects TR values to achieve full T1 relaxation. - T2 times shorten, some metabolites are observed at long TE while others require much shorter TE times of approx. 35 ms. - Decrease in metabolite T2* times: Increased metabolite line widths - a limitation at 3 T vs 1.1 T. - N.B Protocols used must consider the relaxation properties of the metabolites and how the field strength influences them. ## MRS Techniques & Sequences - **A. Single voxel spectroscopy (SVS):** - Obtained from selected voxel acquired from a combination of 3 RF pulses in 3 orthogonal planes - achieved when an RF pulse is applied while a field gradient is switched on. - **B. Multi-voxel MRS: Chemical Shift Imaging (CSI):** - From a single sequence, CSI generates a spectrum for many voxels in different regions of the brain that reflects the chemical shift properties of each metabolite in each individual voxel. - It can use a 2D or 3D technique. - After the application of RF pulse & slice select gradient, PE gradients are applied in 2 or 3 dimensions to sample k-space. FE gradient is not applied. ## SVS vs. CSI - **Single Voxel (SVS)** - Operator set-up: Fast and easy. - Shimming: Limited volume of interest allows very good shim to be obtained. - Spectral quality and peak separation: Excellent with high signal-to-noise, quantifiable. - Spectral contamination: From adjacent tissues due to partial volume and chemical shift displacement effects. - Imaging time: Fast (3-5 min per voxel). - Suitability based on size/characteristics of lesion: Best for medium-sized, homogeneous lesions in large organs. - **Multi-voxel (CSI)** - Operator set-up: A little harder and slower. - Shimming: Difficult to shim well over entire region. - Spectral quality and peak separation: Lower signal-to-noise, problems with quantification. - Spectral contamination: Bleeding of spectra from adjacent voxels due to chemical shift aliasing. - Imaging time: Slower, depends on resolution: 5-8 min for 2D, 7-15 min for 3D. - Suitability based on size/characteristics of lesion: Best for lesions in small organs or for inhomogeneous lesions in larger organs. ## MRS & TE Values - MRS at different TE times results in different spectra. | | Short TE (20-40 ms) | Intermediate TE (135-144 ms) | Long TE (270-288 ms) | |--------------|------------------------------------------------------------------------------------------|--------------------------------------------------------------------------------------------------------------------|--------------------------------------------------------------------------------------| | SNR | High SNR & less signal loss because of minimal dephasing effects. | Reduced SNR - more signal loss because of much more dephasing & decay. | | | Spectrum | Generates a spectrum with more metabolite peaks. | Generates a simpler spectrum with fewer metabolite peaks because of the suppression of some signals. | | | Metabolites | Good for detecting Myoinositol & glutamine-glutamate complex. | easier recognition of lactate Vs lipid. Lactate peak is below baseline, lipid peak remains above. | Good for detecting NAA, Cr & Cho peaks. | ## Brain Metabolites | Spectrum | Abbreviations | Effect | Resonance | |--------------------------------|---------------|---------------------------------------------------------------------------|---------------| | NAA-N-acetyl aspartate. | NAA | Marker of neuronal & axonal viability & integrity | 2.02 ppm | | Lactate (Doublet peak) | Lac | Specific marker of cell death & tissue necrosis | 1.33 ppm | | Choline | Cho | Marker of cell membrane activity & cellular proliferation | 3.22 ppm | | Creatine | Cr-PCr | Marker of cell energy & intracellular metabolism | 3.02 ppm | | Lipids (two peaks) | Lip | Result of cellular decay or necrosis | 0.9; 1.33 ppm | | Myo-inositol | Ins | Glial cell marker | 3.56 ppm | | Glutamine / Glutamate | Glx | Neurotransmitter | 2.05; 2.50 ppm| ## Clinical Applications of MRS - MRS Complement conventional MRI in the assessment of brain tumours to investigate: - Differential diagnosis. - Histological grading. - Degree of infiltration. - Tumour recurrence. - Response to treatment when radiation necrosis is present & indistinguishable from tumour on conventional MRI. - Brain MRS used in neonates to check for development disorders. ## Examples - **Normal:** Voxel in normal left occipital lobe. - **Lesion:** (Patient age 27 with brain mass. Relatively normal spectrum.) - Low-grade glioma. - Similar cellular composition to normal brain tissue. - **Tumour Grading:** (Patient age 47 presents with cognitive difficulties.) - T2W FSE. - CE T1W'SE. - T2W FLAIR. - **High Grade Glioma:** (MRS can be used to determine the grade of a glioma.) ## References - https://www.ndcn.ox.ac.uk/research/mr-spectroscopy - https://www.frontiersin.org/journals/neurology/articles/10.3389/fneur.2019.00482/full - chrome-extension://efaidnbmnnnibpcajpcglclefindmkaj/https://www.thieme-connect.com/products/ejournals/pdf/10.1055/s-0039-1688654.pdf - Bertholdo D, Watchatakorn A & Castillo M (2013). Brain Proton Magnetic Resonance Spectroscopy. Magn Reson Imaging Clinic N Am, 23: 359 - 380. - Tognarelli, M, et al. (2015). Magnetic Resonance Spectroscopy: Principles and Techniques: Lessons for Clinicians. Journal of Clinical and Experimental Hepatology, 5: 320 – 328. - MRI questions & answers website: http://www.mriquestions.com/index.html