Magnetic resonance imaging (MRI) has been increasingly used in the assessment of multiple heart diseases. The versatility of this technique makes it possible to obtain anatomical and functional information, including volume measurement, flow evaluation, myocardial perfusion and tissue characterization.1 When assessing the myocardium, the search for edema and fibrosis assumes particular importance given the diagnostic and prognostic implications.2 The study of myocardial edema is usually performed with T2-weighted images, while the evaluation of fibrosis generally involves the production of T1-weighted images acquired after administration of paramagnetic contrast (late enhancement technique). In both cases, the images obtained translate signal intensity into arbitrary units expressed on a gray scale. This classic approach has three main limitations: first, the analysis3 usually involves visual estimation and is therefore subjective; second, the images are analyzed by making a comparison with an area considered healthy, which may limit its sensitivity in the detection of diffuse changes, especially at an early stage; finally, the late enhancement technique used for fibrosis detection requires the administration of paramagnetic contrast, which is contraindicated in some patients.

Myocardial mapping techniques thus emerge as an alternative to overcome these limitations, providing quantitative information that is potentially useful in the diagnostic and prognostic evaluation of various heart diseases and in understanding the underlying pathophysiological processes. In this narrative review article, we attempt to examine, in a non-exhaustive manner, the evidence supporting the use of T1 myocardial mapping as an add-on to existing MRI techniques. We present some basic technical concepts,4–6 the advantages and limitations, and discuss the potential clinical applications of myocardial T1 mapping.

T1 is estimated by sampling the signal at different points of the exponential recovery curve of the longitudinal magnetization (Mz) after disruption to the equilibrium (Figure 1).7 The inversion-recovery methods are the most widely used because they are available on clinical equipment as a result of their demonstrated high accuracy.7,8 The Look-Locker (LL) method, which enables more than one image after each inversion to be acquired, has been adapted to include synchronization with the electrocardiogram (ECG) in a method called Modified LL (MOLLI).9,10 This process is usually performed in apnea, which can last up to 20 seconds, depending on the heart rate (Figure 2). To reduce acquisition time, and to enable T1-mapping in patients with lower respiratory capacity, a sequence called short MOLLI (ShMOLLI) was developed that reduces apnea to nine cardiac cycles (∼9 s for 60 bpm).11 It is also possible to perform T1 mapping using saturation pulses. Although less accurate than the inversion-recovery methods, the saturation method is more accurate than the previous ones, since each point on the curve is acquired after a new pulse, meaning that samples taken are independent of each other.8 The most commonly used method in this context is SAturation recovery Single Shot Acquisition (SASHA),12 in which ten measurements are taken equidistantly and synchronously with the ECG. Regardless of the advantages and disadvantages of each method, it is important to keep in mind that none of them are able to provide the true myocardial T1 time, but rather provide estimates of this value.

Figure 1.

Standard structure for T1 map acquisition in cardiac MRI. A series of images is acquired (a) and each image has a different T1-weighting. This is possible by changing the acquisition parameter (i.e. inversion time (IT)) for each image, and which corresponds to a point on the exponential recovery curve of the longitudinal magnetization Mz. The scheme represented is MOLLI 5(3)3 composed of two inversions each followed by a set of five or three T1-weighted images, respectively, and separated from each other by three heartbeats (see ECG synchronisation in Figure 2). The images are subsequently reordered (b) according to their respective TI. Finally, (c) for each pixel, the parameters of a known model, including the time T1, are adjusted in order to describe the temporal evolution of the signal. All estimated T1 values are presented in a final T1 map image. Legends: Mz: longitudinal magnetization.

(0,68MB).

Figure 2.

Scheme of synchronisation of the MOLLI method with the ECG. The acquisition represented is MOLLI 3(3)3(3)5 in which three inversions are performed. The first two are followed by three images with different inversion times separated from each other by a heartbeat and the last inversion with five images. Between inversions there is a magnetization recovery period which is represented inside the brackets and which corresponds to three heartbeats. In this way, the acquisition lasts 17 heartbeats, obtaining 11 images in total. These are ordered according to the time elapsed since the respective inversion, in order to perform the adjustment of the exponential model. Legends: ECG: electrocardiogram; Mz: longitudinal magnetization.

(0,15MB).

Much of the early enthusiasm in the development of myocardial mapping was driven by interest in measuring extracellular volume fraction (ECV, sum of intravascular and interstitial components). The expansion of the ECV by interstitial fibrosis is a common process in several myocardial diseases and an important diagnostic and prognostic marker.4,13 Evaluation of the ECV may be particularly useful in the differential diagnosis between physiological and pathological hypertrophy, since in the case of the former, the ECV proportion is not increased (unlike in pathological processes) (Figure 3C). The determination of the ECV fraction using MRI involves the creation of pre-contrast T1 maps (also called “native T1”) and post-contrast (acquired at least 15 minutes after its administration). The ECV is then calculated based on the values of these maps, using the ratio of the differences between post- and pre-contrast (1/T1) myocardial relaxation rates and post- and pre-contrast blood relaxation rates. Calculation of the ECV usually requires knowledge of the hematocrit value since this can influence the transfer of contrast from the blood to the tissues. Alternatively, a theoretical hematocrit value can be used, although the error propagated to the ECV is not insignificant when the hematocrit is too low.14 It is important to realize that the measurement of the ECV fraction is dependent on the applied contrast dose and renal clearance, among other factors.4 Other sources of error include motion (cardiac or respiratory), arrhythmias, or any other factor that impairs the correct overlap of the native and post-contrast T1 maps.

Figure 3.

T1 and VEC maps (short basal, middle, apical and respective polar-plots) in different pathological situations: A - T1 maps of a patient with hypertrophic cardiomyopathy and mean myocardial T1 time of 1085 ms (reference 950-1050 ms); B - T1 maps of a patient with cardiac amyloidosis and mean myocardial T1 time of 1170 ms (reference 950-1050 ms); C - VEC maps of a professional football player with mild left ventricular hypertrophy considered physiological and with mean VEC of 24%.

(0,92MB).

In most clinical contexts, it is still unclear whether and under what circumstances the ECV measurement is superior to native T1, or even whether it provides relevant additional information compared to the latter parameter. Studies have provided conflicting results in this regard.15

A correct evaluation and interpretation of myocardial mapping depends on the existence of reliable reference values. It is important to remember that the normal values of myocardial T1 times are influenced by many factors, including magnetic field strength (1.5 or 3.0 Tesla), as shown in Table 1, as well as the sequence used (which in turn varies according to device manufacturer and even device model).16–21 Other factors that influence the T1 time include heart rate and the region of the heart being assessed (with slightly higher values in the apical segments). Ideally, each center should establish its own reference values using the devices and sequences available to it in healthy volunteers.

Several recent studies have consistently demonstrated a good correlation between histological indices of interstitial fibrosis and native T1, post-contrast T1, and ECV values,23 thus establishing MRI myocardial mapping as a noninvasive tool in the study of fibrosis and interstitial remodeling of the myocardium.24–27 This technique thus increases the ability of MRI to provide useful diagnostic and prognostic information in various scenarios, thus justifying its increasing use in clinical practice.

Despite the evidence presented regarding the usefulness of T1 mapping in each of the diseases under analysis, its actual clinical use is distinct. In diseases such as Anderson-Fabry or amyloidosis (two extremes of the spectrum), the T1 mapping technique has greater potential to be used in diagnosis, due to the greater differentiation between diseased and healthy tissue. However, in settings such as hypertensive heart disease, AS, or HCM, affected tissues may present values close to or even within the reference ranges, with great overlap between diseases. Although the mean values are higher in some diseases than in others, the differences are relatively small, which greatly limits the specificity for a given disease. This limitation means there must be caution in its use and suggests that interpretation should always be made in view of the clinical context and other imaging findings. In this regard, it is important to note that in most clinical situations, T1 mapping complements (and does not replace) the information provided by late enhancement. Pathologies characterized by diffuse interstitial fibrosis will probably be those where this new parameter will assume greater relevance. However, since interstitial fibrosis and replacement fibrosis processes are not mutually exclusive, T1 mapping may also be useful in diseases where localized late enhancement is usually observed, often coexisting with diffuse interstitial fibrosis. In these contexts, it is possible that the main information provided by T1 mapping is principally prognostic rather than diagnostic.

One of the main current limitations of T1 mapping is the relative lack of standardization of the technique. Despite recent progress in this area, the existence of different sequences and the susceptibility of the results to small technical variations mean that there are currently no universal reference values, but only wide ranges, which makes comparability between tests difficult. In fact, as it is a quantitative parameter that is dependent on equipment calibration, equipment characteristics, as well as other intrinsic patient parameters such as field homogeneity at the site of interest, it is a technique that is highly dependent on the environment in which it is applied. A practical consequence of this limitation is that comparative studies should, whenever possible, be performed on the same device and with the same sequence. In addition, obtaining the sequences for mapping requires the purchase of software packages, entailing additional costs that are not accessible to all centers. Access to the necessary post-processing programs is also limited.

Also, from a technical point of view, it is important to mention the reduced spatial resolution currently achievable in T1-weighted maps compared to RT images, which may limit the detail of tissue characterization possible with this technique.

As the technique matures, it is expected that research will focus less on the technical component and more on its clinical applications. Unanswered questions include whether T1 mapping can replace late enhancement in specific situations, its incremental diagnostic and prognostic value relative to existing techniques, and how the results might help guide treatment.

Myocardial T1 mapping is an emerging MRI technique that enables quantitative tissue characterization. Used alongside the usual techniques, T1 mapping enables the evaluation of fibrosis which, when present in a diffuse form, was practically only accessible through myocardial biopsy. In addition, where there is localized fibrosis, it can complement the assessment performed with RT. The information obtained is potentially useful in the diagnosis and prognostic evaluation of several cardiac diseases, especially in the differential diagnosis of left ventricular hypertrophy, in the suspicion of hypertrophic cardiomyopathy or infiltrative diseases, among others. The clinical utility of this tool and its possible implications in the diagnosis and treatment of multiple cardiac pathologies are an active and rapidly expanding area of research.

The authors have no conflicts of interest to declare.