Results: CT Clinical Practice: Data acquisition protocols

CNS CT angiography

1. We have tried performing CT angiogram using Siemens 16 slice scanner. We use spiral, 100 cc nonionic contrast and 15 seconds. (11 seconds plus 4 second scanner delay), but there was very little contrast in the brain. I also tried with axial scan with 25 seconds delay. The contrast was good except I was unable to see the basilar artery. Any recommendations? Also, when doing MIP in 3D, what slice thickness and interval do you use to get the whole COW in 1 or 2 pictures.

Answer: A 17 - 20 second delay is perfect on the 16 slice system, and you get great arterial phase images without venous contamination. We use .75 mm slice thickness at .5 mm intervals. We use VRT on InSpace for the 3D imaging. Let me know if this helps. See some carotids and intracranial vessels at www.insideinspace.com as well as on www.CTisus.com.

2. Your timing for carotids seems long. Do you cover from the arch to the bifurcation? Do you routinely get jugular vein? What do you base this wait time on?

Answer: A 17 second delay works well and we do scan from arch to circle of willis.

3. We are having difficulty getting high resolution vertebral origins (<1 mm) and intracranial vessels in the same run due to limitations in maximum mAs and shoulder artifact. Do you have any suggested protocols other than 2 different runs?

Answer: The key is to use a 16 slice scanner and optimize both pitch and collimation.

4. We are having some problems with brain CTA. We use a single detector helix CT, what do you suggest about slice thickness, delay time, acquisition time, # of rotations, etc? At the most time, we get arterial and venous images together.

Answer: With single slice it is a speed issue. The best I can say is to scan early (15-18 second delay), inject 4 ml/sec and maximize your pitch. The venous contamination is simply a result of the long scan times necessary on a single slice system. When you go to 16 slice, you will see quite a difference.

5. I recently performed a CTA circle of Willis (COW) in a young male using 100 ccs's Visipaque and 50 cc's saline injected at 4.5 cc's per second. I started scanning caudocranially at 25 seconds. I ended up with a substantial amount of venous contamination. I am using the Siemens Sensation 16, so scan time was not much of a factor. Any suggestions to reduce venous contamination? Do you use timing or care bolus for COW? Also, does your protocol change with patient age to account for slower cardiac function?

Answer: A 17 - 20 second delay is perfect on the 16 slice system, and you get great arterial phase images without venous contamination. We routinely use preset timing, and I have found this to be satisfactory. However, bolus tracking, particularly with visual tracking of the bolus, will work as well. In older patients or those with poor cardiac function, we typically add 5 to 10 seconds to the delay. Products such as care bolus are most valuable in this population.

6. In our CT angio protocol using the Siemens Sensation 16 we utilize bolus tracking for CT head and neck angiography. In your protocol list suggestions you utilize a 15 second scan delay. Which in your opinion is better (i.e. better pictures) and is not 15 seconds too long of a delay and what about patients factors regarding cardiac output and circulation time?

Answer: For carotids, most sites use a 15-17 second delay of 100 cc of contrast injected at 3-4 cc/sec. This helps give good arterial imaging without much venous contamination. I like the preset delay, if the patient has a big heart, you can go to a 22-25 second delay.

7. Optimum timing to demonstrate Circle of Willis? How much does this vary depending on the type of scanner (single vs. multislice)?

Answer: It is easiest with 16 slice MDCT. In that case a 16-18 second delay seems ideal.

8. Do you have a jugular bulb protocol for a 16 row GE scanner? Do you recommend a helical or a sequential scan, and what delays would you do? Would you treat it as an angio and do MIPs or Volume rendering if any?

Answer: I would use dual phase with 15 second and 35 second delays. A few would also do a delay at 60 seconds. I would do it as a CT angio with .75 section thickness at .5 mm for the best resolution. VRT will be great and MIP may also prove useful. If you have any luck, send us the case and we will post it.

References


Lell M, Wildeberger JE, Heuschmid M et al. CT angiography of the carotid artery: First results with a novel 16-slice-spiral-CT scanner. ROFO 2002, 174(9): 1165-9.

  • Summary: Using a 16 slice scanner with .75 mm collimation, pitch of 1.5 and table speed of 36 mm/s the authors imaged the carotid arterial system after infusion of 60 ml of contrast and a saline flush. A test bolus was used to dictate timing. The scan time of 9 seconds covered 300 mm, and enabled imaging from the aortic arch through the circle of Willis in the arterial phase. Post processing was performed with MPR and MIP, and the authors noted that pulsation artifact was not an impediment to evaluation of the arteries at the aortic arch level.

Chen C-J, Lee T-S, Hsu H-L et al. Multi-slice CT angiography in diagnosing total versus near occlusions of the internal carotid artery: comparison with catheter angiography. Stroke 2004; 35(1): 83-85.

  • Summary: In this study, catheter angiography and 4 slice multidetector row CT were performed to evaluate 57 internal carotid near occlusions or total occlusions. The CT acquisition was initiated 20 seconds after the start of contrast infusion (120 mm at 3 cc/sec), using 1.25 mm section thickness and 7.5 mm per rotation table speed. Curved planar reconstructions and MIP renderings were performed. There were 31 total occlusions and 26 near occlusions. CT angiography accurately diagnosed all occlusions, near or total (100% sensitivity and specificity), as well as the stump measurement and any retrograde flow (100% correlation with catheter angiography).

Lawler LP, Corl FM and Fishman EK. Multi-detector row and volume-rendered CT of the normal and accessory flow pathways of the thoracic systemic and pulmonary veins. Radiographics 2002; 22: S45-S60.

  • Summary: This article includes images displaying systemic and pulmonary veins with 4 slice MDCT and volume rendering. A discussion of acquisition timing tailored to the region of interest is provided. The sample protocol for this type of scanner included 1 mm collimation (2.5 mm in those who can not hold their breath), 1.25 mm section thickness, 1 mm reconstruction interval. A case of jugular vein occlusion is displayed.