Nicholas C. Gourtsoyiannis - Clinical MRI of the Abdomen: Why,How,When

(1)

(2)

Nickolas Papanikolaou

Email:

Magnetic resonance imaging (MRI) is considered as one of the most important imaging modalities that is steadily growing its contribution to the routine diagnostic workup of abdominal diseases. Among the absence of radiation exposure and the rich soft-tissue contrast, MRI has inherent three-dimensional imaging capabilities providing images in all three orthogonal planes, as well as in oblique or even double oblique orientations. Due to its comprehensive imaging nature, its possible to explore not only morphology, but function and physiology based on novel contrast mechanisms like molecular water diffusion, tissue perfusion, and MR spectroscopy.

The introduction of physiology-based MRI techniques have upgraded the role of MRI recently. These new sources of information may help to improve tissue characterization and perform more accurate diagnosis, as well as be used for monitoring the response to treatment. In the era of individualized medicine, MRI may play an important role to guide noninvasively novel therapeutic approaches.

Approximately, 70% of the human body is water. The number of hydrogen nuclei that can be found on water molecules in human body can be as high as 5 1027. Each individual hydrogen nucleus is described by a magnetic momentum (spin); therefore, it can be considered as a small magnet. In the absence of external magnetic field, spins have arbitrary direction. When the patient is entering the magnetic field his spins is forced to become parallel or antiparallel to the direction of the external magnetic field (Bo). The sum vector of all these small magnetic momentums lies parallel to the external magnetic field, while its magnitude is directly related to the strength of the external magnetic field. The acquisition of an MR image comprises different phases, the most important of which are: the excitation, the relaxation, and final signal reception and image reconstruction. The MRI signal is generated from amplitude changes of the transversal magnetization vector of spins that belongs to a specific two- dimensional slice. This can be done by a dedicated magnetic field called gradient field. To spatially encode data that are coming from a specific slice level such a gradient field can be recruited. The basis of spatial encoding is to spatially vary the strength of the static magnetic field, in such a way where only the spins that belong to a specific slice resonate exactly at the resonance frequency of hydrogen. In this way, the signals that receiver coils will acquire will come only from the selected slice. In a similar way, two other gradient fields can be incorporated to spatially encode the x and y coordinates on a two-dimensional image.

Magnetic resonance imaging is based on the interaction between a varying magnetic field (excitation termed as B1) and the hydrogen nuclei (single protons) that can be found in the form of water in the human tissues, in the presence of a strong static magnetic field (termed as Bo). The image contrast depends on proton density as well as on the way that spins react to the initial excitation. The latter can be described by two different time constants that are called T1 and T2 relaxation times. The value of T1 and T2 relaxation times depends strongly on tissue composition. More specifically, tissues with high water content presents with long T1 and T2 relaxation times, while the opposite is true for solid type of tissues. This might explain why MRI can provide excellent soft tissue contrast. By adjusting key parameters in our sequences, it is possible to acquire images where contrast is based on differences regarding proton density and T1 and T2 relaxation time values.

Recent technical advances in the field of hardware and software made new applications feasible. The most important one among them maybe the diffusion-weighted imaging. This technique is well known from its applications in the brain to study acute ischemia and tumors [). So, increased diffusion of the water molecules can be found in most of the benign lesions due to the destruction of tissue architecture and presence of edema, while increased cellularity is responsible for restricted diffusion that is a hallmark of malignant lesions.

Technical limitations in gradient systems are responsible for the contamination of diffusion images contrast from other mechanisms like T2 or T1 relaxation. To minimize these effects, the shortest available echo time should be selected on the diffusion pulse sequence. In the setting of a weak gradient system, the minimal echo time that can be achieved may still be too long and therefore significant T2 contamination may take place. In order to avoid the so-called T2 shine through effects (based on T2 relaxation contamination), an alternative strategy includes the reconstruction of apparent diffusion coefficient maps that are parametric images and contain pure quantitative diffusion information. The apparent diffusion coefficient (ADC) is a physical property of each individual tissue and its value is mainly affected mostly by the size of extracellular space. In these ADC maps, benign lesions that generally result in the enlargement of extracellular space due to increased water content, which presents as bright areas (high ADC values), whereas malignant lesions commonly expressed by hypercellularity result in shrinkage of extracellular space and present as dark areas (low ADC values). One of the most important challenges for abdominal applications of diffusion is physiological motion. There are two different approaches to overcome this problem, the utilization of either breath-holding or respiratory triggering techniques. In case of breath holding, the resulting diffusion images suffer from low spatial resolution. Respiratory-triggered diffusion sequences may reduce respiratory-related artifacts, in the setting of stable respiratory cycle.