Intravenous coronary angiography with electron beam computed tomography (EBCT) allows for the non-invasive visualisation of coronary arteries. With dedicated computer hardware and software, three dimensional renderings of the coronary arteries can be constructed, starting from the individual transaxial tomograms. This article describes image acquisition, postprocessing techniques, and the results of clinical studies. EBCT coronary angiography is a promising coronary artery imaging technique. Currently it is a reasonably robust technique for the visualisation and assessment of the left main and left anterior descending coronary artery. The right and circumflex coronary arteries can be visualised less consistently. Improvements in image acquisition and postprocessing techniques are expected to improve visualisation and diagnostic accuracy of the technique.
- electron beam computed tomography
- coronary angiography
- coronary anatomy
- volume rendering
Statistics from Altmetric.com
Recently intravenous coronary angiography using electron beam computed tomography (EBCT) has emerged as a new coronary imaging technique.1-7 By virtue of the absence of moving parts, EBCT can construct ECG triggered, high resolution tomograms in 100 ms. This is fast enough to reduce cardiac motion artefacts and produce high quality cardiac images,8 especially when the acquisition window is set during diastasis. The scanner was originally designed in the 1980s for cardiac function studies and myocardial perfusion imaging.8 Later it was applied for the accurate quantification of coronary calcification.9
Scanning is performed with the patient in the supine position. Contrast (iopromide 350 mg/ml) is injected through an antecubital vein to a total of 150 ml. Tomography starts just proximal to the left main coronary artery after an ECG trigger at 80% of the RR interval (diastasis). A tomogram thickness of 1.5 mm and a table increment after each tomogram of 1.5 mm results in contiguous non-overlapping slices. A total of 40–60 contrast enhanced tomograms are made during a single breath hold after intravenous contrast administration (fig 1). Radiation exposure for a complete examination is estimated to be < 20 mGy.3 9 This is approximately one fifth of the radiation exposure at coronary angiography.10 It is our3 and others’11 experience that after instruction most patients are able to hold their breath for at least 35 seconds.
The individual tomograms have a resolution of 4–6 linepairs/cm and the scanning time for each tomogram is 100 ms. The spatial resolution in the scanning direction is limited by the slice thickness of 1.5 mm. Very short lesions (< 3 mm long) in artery segments that are perpendicular to the scanning plane can therefore be missed. Although these specifications fall short of the spatial (5 linepairs/mm) and temporal resolution (up to 50 frames/s) of conventional cine coronary angiography, proximal and middle parts of the coronary arteries can be visualised with EBCT.
THREE DIMENSIONAL RECONSTRUCTION
Two dimensional transaxial tomograms are transferred to a computer workstation where they are stacked and interpolated to form a three dimensional volume using specialised software. This process creates a database representation of a three dimensional object. With special rendering techniques this database representation of the three dimensional object can be visualised. Several three dimensional rendering techniques are available,12 but we almost exclusively use volume rendering.3
Volume rendering allows certain parts of the three dimensional object to be transparent so that structures behind these parts are still visible. By assigning the right colour and opacity values to different tissues, real anatomy is simulated in much the same way as pictures in an anatomic atlas (figs 2, 3, and 4).13
Several groups have compared intravenous EBCT coronary angiography to conventional angiography.2-7 In all these studies both the original tomographic data and the three dimensional renderings were used to assess the coronary arteries. Overall, the technique allowed visualisation of the proximal and mid- coronary arteries in approximately 80% of cases. The left main coronary artery and proximal and middle parts of the left anterior descending coronary artery could generally be visualised and assessed in 90–100% of cases, with a sensitivity to detect a significant stenosis of 85–90% and a specificity of around 90%. Images of the right and circumflex coronary arteries were interpretable in only 75% of cases, with a little lower diagnostic accuracy in general. Major causes for non-assessability were cardiac motion artefacts, and the smaller size of the middle and distal parts of the right and circumflex arteries. Improvements in ECG triggering, shorter acquisition time (50 ms high resolution tomograms), and improved spatial resolution with a recently introduced new detector ring are expected to improve visualisation and diagnostic accuracy of the technique.
Intravenous EBCT coronary angiography is a promising imaging modality that allows for non-invasive visualisation of the proximal and middle parts of the coronary arteries.