A single tube capable of switching rapidly between the low and high tube potential.

This approach trusts on the ability of the detector to differentiate low-energy from high-energy X-ray photons in a single X-ray beam. The tube potential switching occurs as rapidly as every 0.5 ms. Each pair of 80 and 140 kV projections are essentially acquired from the same view angle. To avoid spectral contamination between consecutive 80 and 140 kV projections, the scanner uses a scintillating material (gemstone) that has an ultrafast primary falloff time (0.03 ms) and low afterglow (delayed fluorescence) (Fig. 12.2).

There is minimal view angle mismatch between the successive high and low tube potential projections so material decomposition can be performed in the projection space which would lead to a more exact BH correction. Table 12.2 summarizes the technical configuration.

A single-source dual-layer detector system (Philips Medical Systems) more closely resembles the ideal energy-resolving photon counting system (Fig. 12.3).

The x-ray detector consists of two different scintillating materials bonded together. This design would allow higher energy X-ray photons to pass through the top layer without suffering significant interaction, whereas the lower energy photons are mostly attenuated in the top layer.

The two signals, one from the top and one from the bottom of this sandwich detector, would correspond to x-rays in two different energy ranges. Table 12.3 epitomizes the technical settings.

The main advantage of this design is that at each view the high- and low-energy projections are exactly registered with respect to each other.

However, one potential problem is that the high- and low-energy projections are spectrally less separated from each other compared with the rapid tube potential switching or dual-source system, which could lead to suboptimal material decomposition and BH correction for quantitative MP imaging.

In general, a triphasic protocol is selected. Ideally an 18 gouge access is chosen and placed in the right antecubital vein. It is connected to a dual-syringe injector.

The delay time can be determined by a test bolus using 20 ml of non-ionic contrast and 30 ml of saline solution in a dual phase injection. The preferred injection rate is 5–6 ml/s, if possible.

Automatic bolus tracking is another option. It requires placing a region of interest (ROI) in the ascending aorta with a predefined threshold of 130 UH. When the contrast reaches that level the scan is initiated automatically.

The triple phase injection consists of a first phase of undiluted contrast (60 ml), a second phase of mixed contrast and saline solution (60 ml: 30 %/70 %) and a third phase of pure saline solution (30 ml).

Images obtained are initially called “mixed images” that means that are obtained by linear weighting of the CT value for the two spectra. These images are equivalent to the standard CT ones and correspond to an intermediate energy.

Once acquired, images can be decomposed in different energy levels. The spectrum ranges from 40 to 140 keV.

Each level will allow visualizing the myocardium with higher and lower iodine concentration.

Low energy levels show higher contrast enhancement in the myocardium as well as in the coronary vessels but are associated with increase in noise. In this level also beam hardening artifact can be present. On the contrary, higher energy levels have lower myocardium and intravessel contrast with lower image noise and beam hardening artifact is cancelled (Fig. 12.4).

Images at 140 keV will be similar to non–contrast images. They are known as “virtual non contrast images”. For myocardial perfusion these levels are not useful. However, they are advantageous for the quantification of calcium score from a dual energy scan with contrast load. Traditional contrast calcium score acquisition before the CCTA study can be therefore avoided.

In monochromatic grades, images can be analyzed without beam hardening artifact. In general, the best levels for the assessment of myocardial perfusion are those between 70 keV or greater (Fig. 12.5).

BHA is a limitation of Single Energy CTP, which can demonstrate difficulties to confirm or rule out perfusion defects in cases with moderate CAS with associated perfusion defects in the corresponding coronary territory. In those situations it is difficult to determine if such perfusion defect is a true positive or false negative finding. DECT may play an interesting role in these scenarios.

DECT permits to identify any perfusion defect and evaluate its behavior among the different energetic levels. The evaluation can be qualitative or quantitative. Grey scale and color scale are available forth analysis.

In cases with a true perfusion defect, it will persist in all energetic levels. However, if there is a BHA, the pseudo-hypoperfusion will be present only in low energetic levels (40 keV), it is going to attenuate in higher levels (from 40 to 70 keV) and will disappear in levels above 100 (Figs. 12.6 and 12.7) [63].

In the stress scan, systolic and diastolic phases are required for the analysis, as motion artifacts can limit the accuracy.

Images are reprocessed in short axis views and complemented in long axis in order to perform the 17 segmental analysis of the American Heart Association (AHA).

DECT short axis images are best evaluated with slice thickness of 4–5 mm instead of thicker 8 mm-slices that are suggested for SECT.

Once the best energy level is determined a quantitative analysis can be complemented in a similar manner to SECT. Measurements in HU can be done in each of the 17 segments in the subendocardial layer as well as in the subepicardial layer and obtain the transmural index plus mean HU value in normal vs. abnormal segments.

This type of analysis is based in the main principle of DECT.

DECT enables a better assessment of the different tissues and materials that can be found in the body without artifacts such as beam hardening.

Material decomposition is based on different attenuation coefficients, which depend on the energy levels of the X-ray beam.

It also gives information about effective atomic number and mass density maps, in the form of basic materials such as water or iodine.

Using material decomposition is possible to select two different materials in an arbitrary manner being able to suppress one and select the other one.

Also it is possible to quantify the amount of each material per mm3 of tissue. The amount can be given in ug, ng or mg per mm3 of tissue.

For myocardial perfusion a useful combination is iodine/water.

This type of analysis permits to measure the amount of iodine per mm3 of the myocardium and determine if there is a normal or an abnormal perfusion. Iodine distribution in the myocardium can be a surrogate of myocardial blood pool.

Figures 12.8 and 12.9 are examples of material decomposition in normal and pathologic iodine maps of the myocardium respectively.

Material decomposition can be shown in gray scale or in color scale. There are several options for color coding of the iodine map such as hot iron, Page phase, rainbow, dark gold·D, French and others (Fig. 12.10).

Several authors such as Schwarz et al. [54]. prefer “Micro-Delta-Hot-Metal” color-coding, because they believe it most closely resembles the SPECT images from myocardial perfusion studies.

DECT has shown good results to identify reversible perfusion defects, since by means of material decomposition and/or monochromatic evaluation a more precise information can be obtained.

One of the first studies was performed by Ruzscis et al. [52], including 36 patients with equivocal or discordant SPECT. DECT studies were carried out with first generation DSCT. There were 157 hypoperfused segments on SPECT, 89 were fixed. DECT correctly identified 96 % of fixed and 88 % of ischemic myocardial perfusion defects. Overall sensitivity and specificity for any defect was 92 % and 93 % respectively.

Another paper published by Wang et al. [64] done in 34 patients with known CAD or pathologic SPECT evaluated the usefulness of rest DECT versus ICA as gold standard modality. Thirty one out of 34 patients achieved adequate image quality. DECT had a sensitivity of 68 % and specificity of 93 % for the detection of any myocardial perfusion defect. Combined anatomic and functional evaluation showed small improvements in diagnostic accuracy (from 86 % to 88 %).

In the paper of Weininger et al. [51] “Adenosine-stress dynamic real-time myocardial perfusion CT and adenosine-stress first-pass dual-energy myocardial perfusion CT for the assessment of acute chest pain”, the authors evaluated dynamic CT and DECT for myocardial perfusion assessment in patients with acute chest pain. The population studied were 20 consecutive patients (15 men, 5 women; mean age 65 ± 8 years) who presented with acute chest pain and who had a clinically indicated stress/rest SPECT and cardiac MRI. Patients were divided into two groups: group A (n = 10 patients) underwent adenosine-stress dynamic real-time myocardial perfusion CT and group B (n = 10 patients) adenosine-stress first-pass dual-energy myocardial perfusion CT using the same CT scanner in dual-energy mode. Two independent observers analyzed all CT perfusion studies in a qualitative manner. CT findings were compared with MRI and SPECT.