Blood Flow Velocity Assessment

Time-of-Flight Methods

There are two categories of TOF techniques. The first, often known as wash-in/wash-out, or flow enhancement, normally rely on the saturation or partial saturation of material in a selected slice or volume being replaced by fully magnetized “high signal” spins as a result of flow ( Fig. 6.1A ). The second involves some form of tagging and then imaging to follow the motion of the tagged material ( Fig. 6.1B ).

Singer and Crooks adopted the first approach in an attempt to measure flow in the internal jugular veins, although quantification was questionable because of other factors affecting the flow signal. The first report to describe a tagged TOF approach was by Feinberg and colleagues. Their method involved a variation on a dual echo spin echo sequence; the first 180-degree selected slice was displaced by 3 mm from the initial excitation slice, and the second was displaced by 9 mm. The first 180-degree selection overlapped sufficiently with the 90-degree selection to produce a good anatomic image. The second 180-degree pulse selection did not overlap with the 90-degree or the first 180-degree pulse selection and therefore produced no anatomic image but gave high signal from the blood that had experienced all of the preceding radiofrequency (RF) pulses (i.e., that had passed between the different selected planes). The technique was used to identify flow in the carotid and vertebral arteries of a volunteer's neck, although flow velocities had to be within a specific range and could not be defined accurately.

Methods have also been described in which the TOF flow movement can be visualized directly on an image. The methods involved the application of slice selection and frequency encoding in the same axis. In this way, material that had moved in this axis between selection and readout would be displaced relative to the stationary material. These techniques were therefore making use of signal misregistration, an effect that is often seen as a problem in other methods of flow imaging. Another TOF approach is to saturate a band of tissue, for example, in a transverse plane, and then to follow the progress of this dark band in the coronal or sagittal plane. The major limitation of these saturation methods is that they are limited by the T1 of the various tissues being saturated. The contrast of the saturated blood will decrease with time, eventually making it difficult to measure accurately the distances traveled. Also, motion during the sampling gradients results in signal and thus image distortion. In addition, for arterial flow measurements in which cardiac gating is required, only two-dimensional (2D) images can be acquired in a reasonable time so that only limited details of the flow profile can be studied.

Phase Flow Imaging Methods

Considerable knowledge had been gained on the measurement of flow from the phase of the cardiac magnetic resonance signal from the nonimaging studies after an original suggestion by Hahn in 1960 of a method of measuring the slow flow of currents in the sea. In 1984, the first attempts of measuring blood flow were described by van Dijk and Bryant and associates, using methods based on the theory suggested by Moran 2 years earlier. The imaging methods that followed fell broadly into two categories:

• 1.

  Phase contrast velocity mapping methods that mapped the phase of the signal directly to measure the flow.

• 2.

  Fourier flow imaging methods that phase encoded flow velocity to produce an image after Fourier transformation with velocity resolved on one image axis.

Both of these methods rely on the same principles that cause flowing material to attain a phase shift that is related to its motion. Fig. 6.2 shows these principles for a fluid flowing down a tube surrounded by stationary material. A bipolar gradient pulse is applied, consisting of a positive magnetic field gradient, followed a certain time later by an equal but opposite negative magnetic field gradient in the direction of the flow. During the period of the positive gradient, flowing and stationary materials in a particular location will take up a frequency shift that depends on their position in the direction of the field gradient. When the gradient is turned off, the phase of the flowing and stationary materials can be considered equal. In the period between the positive and negative gradients, the flowing material moves away from its stationary neighbor. During the period of the negative gradient, the stationary material takes up an equal but opposite frequency shift and returns to the phase it had before the first gradient. However, the flowing fluid takes up a different frequency shift that is dependent on the distance it has moved; its final phase therefore also depends on this distance and hence its velocity.

The relationship between the phase of the signal and flow velocity is:

where A _g is the area of one gradient pulse (amplitude × duration), Δ is the time between the centers of the two gradient pulses, v is the velocity, and γ is the gyromagnetic ratio. A _g , Δ, and γ are all constants for a particular imaging sequence so that a quantitative measure of velocity can be determined if the phase shift can be measured.

The two principal approaches of using the phase shift to produce a quantitative flow image, phase contrast velocity mapping and Fourier flow imaging, are discussed in this chapter.