Characterizing cardiac tissue using MRI is a desirable goal that with the advances of modern scanning technology is becoming ever more possible. One approach presently revolves around attempting to obtain images of the heart that are maps of specific NMR relaxation parameters (typically T2*, T2, T1) and then relating the changes in these fundamental properties to disease in the tissue so as to decide on treatments, these fundamental parameters can be enhanced with exogenous or intrinsic contrast agents. Several acquisition approaches have been demonstrated and these methods, their clinical utility and the underlying mechanisms will be discussed.
Characterizing disease with NMR relaxation parameters has been a theme from the earliest beginnings of the field [Damadian 1971]. The argument for using relaxation parameters for characterizing tissues, is rather simplistic and goes like this: “NMR relaxation parameters are sensitive to the way that water interacts with the environment in which the water exists, proteins can influence the relaxation and we know that fluids have longer relaxation times than more cellular materials. In disease tissues change and the water environment will change, therefore we can detect disease using relaxation parameters, further perhaps we can differentially characterize disease using this method.” In practice, predicting how relaxation times might change in specific diseases has not been easy, but the clinical approach of empirically determining the changes and then working out why seems to have been quite effective in moving the field forwards.
With the emergence of imaging combined with relaxation measurements it is possible to build a relaxation map, which is an image where at each pixel the relaxation value is shown. These are particularly helpful and are an essential step in the take-up of these techniques in the clinic. In the heart this approach has taken a particularly long time to make it into clinical use owing to the initial challenges of obtaining high quality reliable images of the heart, and subsequently with getting multiple images of the heart that are identical except for having a different sensitivity to an NMR relaxation parameter. With the emergence over the last 15 years of high performance gradient systems, the invention of parallel imaging and the availability of fast reconstruction computers now it is very straightforward to acquire maps of T1, T2 or T2* within a single short breath-hold on a standard clinical scanner made by any of the major vendors.
This talk discusses the methods and utility of these three primary contrast mechanisms, and then discusses other sources of relaxation information.
T2*
T2* is perhaps the simplest of the relaxation parameters in cardiac MRI. As is well know the T2* is the relaxation time constant that is measured in a gradient echo acquisition, that is it is the way the signal decays as the nuclear spins diphase relative to one another. Therefore if the tissue/sample has a range of magnetic fields within it then the spins will precess at different frequencies and this will result in more rapid dephasing and a shortening of the T2*. T2* is therefore sensitive to both macro and micro scaled magnetic field gradients. Note that these field gradients are due to the tissues being “magnetized” by the static magnetic field and so 1/T2* increases approximately linearly with the static magnetic field.
Macroscopic magnetic field gradients are caused by magnetic susceptibility interfaces between tissues (for example the heart/lung boundary) and though we can attempt to mitigate these through the use of shimming, in practice in some regions of the heart a uniform magnetic field can never be achieved, hence these macroscopic gradients place a limit on how accurately we can measure long T2* values.
Microscopic magnetic field gradients are considerably more useful for tissue characterization. There are two common mechanisms for these gradients: iron and oxygen, on both cases these agents will be the source of gradients that will influence T2*. Changes in the magnetic properties of Oxygen (the BOLD effect, Blood Oxygenation Level Dependent) can also be see in the heart, but owing to the macroscopic magnetic field gradients and changes in these from breathing and the cardiac cycle can be a difficult thing to determine. Iron has been proven to be useful clinically as it has been shown that iron can be accurately measured using standard T2* measurements of the heart using a multiple echo gradient echo acquisition [Wood 2007]. Iron and oxygenation also affect the T2, but as T2* is easier to measure than T2 on a clinical scanner in a short breath-hold the T2* approach has become very popular for iron measurements. The clinical application of iron measurement in the heart is in diagnosis and monitoring of iron overload and its treatment that is due to thalassemia and other iron loading diseases.
T1
The T1 or spin-lattice relaxation time describes the exponential recovery of the spin system to thermal equilibrium. Nuclei are contained within a molecular structure, and are in constant vibrational and rotational motion, creating a complex magnetic field. The magnetic field caused by thermal motion of nuclei within the lattice is called the lattice field. The lattice field of a nucleus in a lower energy state can interact with nuclei in a higher energy state, causing the energy of the higher energy state to distribute itself between the two nuclei. Therefore, the energy gained by nuclei from the RF pulse is dissipated as increased vibration and rotation within the lattice, which can slightly increase the temperature of the sample. The name spin-lattice relaxation refers to the process in which the spins give the energy they obtained from the RF pulse back to the surrounding lattice, thereby restoring their equilibrium state.
In-vivo T1 is reduce in tissues due to the waters interaction with proteins. Fluids e.g. Cerebrospinal Fluid, have long T1’s owing to very few interactions of this sort. In practice the range of T1’s invivo are quite narrow, perhaps ~500ms for intracellular water and ~1500ms for extra cellular water in the myocardium. As water exchanges rapidly between these different spaces the water may observe both environments hence, the measured T1 is a weighted sum of the T1 values. As a consequence the measured T1 is a complex measure of the T1 in each spaces weighted by the water volume in each of these spaces. For many cardiac disease the T1 ends up being a measure of the extra-cellular space, for some conditions (i.e. Fabry’s disease) the relaxation properties are actually affected, and in others such as iron overload it is the iron that directly influences the T1. Several different approaches have been proposed for measuring the T1 on clinical scanners.
Other variants of these methods have been presented including variations on how the data are fit on a pixel by pixel basis and other combinations and permutations of inversion and saturation pulses.
In the clinical T1 mapping has been found to have real utility across a range of cardiac diseases (see review) and has been standardised by the SCMR [Moon 2013] to help its clinical adoption. These methods are supported by all the major MRI vendors and these approaches are finding their ways into clinical guidelines [Bulluck 2015]. In concert with contrast agents T1 mapping provides quantitative information on the contrast agent concentration and in a situation where the contrast agent is in equilibrium in the extracellular fluid it is possible to quantitatively determine the extra cellular volume fraction in the heart. The ECV may provide additional complementary information to the native T1 [Bulluck 2015].
T2
The T2 or spin-spin relaxation time has similar properties to T1. The relaxation mechanism is different to spin-lattice relaxation and to T2* relaxation in that it represents dephasing of the spin system (like T2*) but is dephasing that cannot be refocused by a 180degree refocusing pulse. This is a complicated principle but the critical difference to T2* is that a static magnetic field gradient will be refocused by a 180degree pulse and so T2 is less sensitive to macroscopic magnetic fields. T2 is therefore sensitive to both microscopic magnetic fields and to the random motion of the water within these environments. In practice, T2 demonstrates very similar characteristics in the heart to increased extracellular volume fraction and to these contrast mechanisms may be in some ways interchangeable.
Presently the dominant T2 mapping approach uses a cardiac gated snapshot balance SSFP acquisition preceded by a T2 preparation module. This approach was pioneered by Giri and Simonetti [Giri 2009] and has been used quite widely. At present the image quality from T2 mapping is inferior to the T1 maps available, but the fractional changes in T2 are larger than the fractional changes in T1 with disease. T2 therefore has great promise as a contrast mechanism and may benefit from more advanced reconstruction methods that may enable fast-spin-echo/CPMG based acquisition [Altbach 2005].
Other relaxation mechanisms and the future
T1rho, magnetization transfer, have also been shown to change in the presence of cardiac infarction and continue to be developed and advanced. Proton density may also have a role to play in disease diagnosis and whilst this is not a relaxation parameter it will be difficult to extract without eliminating the influence of relaxation parameters. Further, the up and coming MR Fingerprinting methods attempt to determine multiple relaxation parameters in a single acquisition and so are an exciting evolution of this field. There is little doubt that Relaxation mapping of the heart will have an important role in the clinic in the future in the accurate stratification of patients for treatment.Damadian 1971: Damadian R. Tumor detection by nuclear magnetic resonance. Science. 1971 Mar 19;171(3976):1151-3. Wood 2007: Wood JC. Magnetic resonance imaging measurement of iron overload. Current opinion in hematology. 2007;14(3):183-190.
Messroghli 2006: Messroghli DR, Radjenovic A, Kozerke S, Higgins DM, Sivananthan MU, Ridgway JP. Modified Look-Locker inversion recovery (MOLLI) for high-resolution T1 mapping of the heart. Magn Reson Med. 2004 Jul;52(1):141-6.
Piechnik 2010: Piechnik SK, Ferreira VM, Dall'Armellina E, Cochlin LE, Greiser A, Neubauer S, Robson MD. Shortened Modified Look-Locker Inversion recovery (ShMOLLI) for clinical myocardial T1-mapping at 1.5 and 3 T within a 9 heartbeat breathhold. J Cardiovasc Magn Reson. 2010 Nov 19;12:69. doi: 10.1186/1532-429X-12-69.
Roujol S 2014: Roujol S, Weingärtner S, Foppa M, Chow K, Kawaji K, Ngo LH, Kellman P, Manning WJ, Thompson RB, Nezafat R. Accuracy, precision, and reproducibility of four T1 mapping sequences: a head-to-head comparison of MOLLI, ShMOLLI, SASHA, and SAPPHIRE. Radiology. 2014 Sep;272(3):683-9
Chow 2014: Chow K, Flewitt JA, Green JD, Pagano JJ, Friedrich MG, Thompson RB. Saturation recovery single-shot acquisition (SASHA) for myocardial T(1) mapping. Magn Reson Med. 2014 Jun;71(6):2082-95.
Moon 2013: Moon JC, Messroghli DR, Kellman P, Piechnik SK, Robson MD, Ugander M, Gatehouse PD, Arai AE, Friedrich MG, Neubauer S, Schulz-Menger J, Schelbert EB; Myocardial T1 mapping and extracellular volume quantification: a Society for Cardiovascular Magnetic Resonance (SCMR) and CMR Working Group of the European Society of Cardiology consensus statement. J Cardiovasc Magn Reson. 2013 Oct 14;15:92. doi: 10.1186/1532-429X-15-92.
Bulluck 2010:Bulluck H, Maestrini V, Rosmini S, Abdel-Gadir A, Treibel TA, Castelletti S, Bucciarelli-Ducci C, Manisty C, Moon JC. Myocardial T1 mapping. Circ J. 2015;79(3):487-94
Altbach 2005:Altbach MI1, Bilgin A, Li Z, Clarkson EW, Trouard TP, Gmitro AF. Processing of radial fast spin-echo data for obtaining T2 estimates from a single k-space data set. Magn Reson Med. 2005 Sep;54(3):549-59.
Giri 2009: Giri S, Chung YC, Merchant A, Mihai G, Rajagopalan S, Raman SV, Simonetti OP. T2 quantification for improved detection of myocardial edema. J Cardiovasc Magn Reson. 2009 Dec 30;11:56