Single rotation CTA of extracranial carotids integrated with cerebral CTP provides sufficient quality for decision making in patients with ischaemic stroke

Abstract

Purpose: Large volume computed tomography scanners with 16 cm Z-axis single rotation coverage enable joggle-mode scanning of cerebral computed tomography perfusion and single rotation computed tomography angiography of cervical arteries. Our study aims to evaluate the feasibility of scanning cervical arteries, acquired with single rotation computed tomography angiography during computed tomography perfusion in ischaemic stroke patients.

Materials and methods: A total of 143 patients were scanned with a single contrast medium injection of 60 ml. Hounsfield units of the cervical arteries and veins were objectively measured and carotid bifurcations were subjectively reviewed. The incidence of artefacts and supra-aortic vessel coverage was recorded.

Results: Single rotation computed tomography angiography of the neck demonstrated supra-aortic vessels to their origins in 58 (40.6%) patients. Ninety-nine per cent (1140/1152) of arterial segments were adequately opacified (≥150 Hounsfield units). Arteries were adequately contrasted compared to veins in 81.3% (915/1126) of segments. However, the opacification was reversed in 14.0% (158/1126) of segments, indicating a delayed timing of acquisition; 95.5% (273/286) of carotid bifurcations were of good image quality. Measurement of internal carotid artery stenosis in single rotation computed tomography angiography according to the North American Symptomatic Carotid Endarterectomy Trial correlated well with digital subtraction angiography (R=0.87, P<0.05). Significant artefacts resulted from metal/dental implants (10.5%), contrast in central veins (7.7%) and the shoulder region (4.9%).

Conclusion: Single rotation computed tomography angiography of the neck incorporated into cerebral computed tomography perfusion with single contrast medium administration revealed adequate image quality for further decision-making in our patient sample. The main drawbacks were inadequate coverage of supra-aortic arteries and possible delay in timing of the joggle.

Keywords: Acute stroke imaging; computed tomography angiography; extracranial carotid arteries; joggle mode CT; one stop imaging; volume CT scanner.

Figure 1.

Plot chart of the measured average Hounsfield units (HUs), minimum HUs and maximal HUs of the arteries and veins. Overview of intraluminal contrast densities in HUs, measured in the predetermined arterial (ascending aorta and carotid arteries) and venous (superior vena cava (SVC) and jugular veins) segments. Note the relatively higher contrast densities in the distal common carotid artery (CCA) and proximal internal carotid artery (ICA) (segments around the carotid bifurcations). This is advantageous during image analysis due to the possible presence of stenosis in the carotid bifurcation.

Figure 2.

Pseudo-thrombus. Single rotation computed tomography angiography (srCTA) images of carotid bifurcations in a 66-year-old patient (a, coronal) and 70-year-old patient (b, curved reconstructed coronal). In (a), the average Hounsfield units (HUs) of intraluminal contrast measured approximately 500, while in (b), the average HUs were approximately 200. The maximal HUs of calcified wall plaques in both images measured 1200. Both images show streak artefacts (white arrow in (a) and black arrow with white outline in (b)) from the calcified wall plaques cast into the lumen of carotid arteries. The artefacts are, however, darker in (b) because of less intraluminal contrast, and these may mimic thrombus.

Figure 3.

Scatterplot chart. Graphical representation of the degree of proximal internal carotid artery (ICA) stenosis (following North American Symptomatic Carotid Endarterectomy Trial (NASCET) criteria) in single rotation computed tomography angiography (srCTA) plotted against digital subtraction angiography (DSA) showing a high degree of correlation.

Figure 4.

Pseudo-occlusion. A 65-year-old patient had right middle cerebral artery (MCA) occlusion (not shown). (a) Coronal section and (b) axial section showing single rotation computed tomography angiography (srCTA) images with circumferential severe wall thickening of the right carotid bifurcation (black arrows with white outline) causing high-grade stenosis in the right internal carotid artery (ICA) origin. (c) Axial section just downstream from the right ICA stenosis showed no contrast in the right proximal ICA (black arrows) while the right external carotid artery (ECA) (white arrow) remained opacified, suggestive of occlusion of the right ICA downstream of a high-grade stenosis at its origin. Nevertheless, right carotid digital subtraction angiography (DSA) (d) showed contrast flowing through the near complete stenosis at the right ICA origin (black arrow) and opacified the downstream ICA.

Figure 5.

Single energy metal artifact reduction (SEMAR). Dense vertebral prosthesis (Hounsfield units (Hus) max 2995) present in a 58-year-old patient. Computed tomography scanning with SEMAR. (a) Axial section and (b) curved coronal reconstruction, showing streak artefacts around the prosthesis. However, the right ((white arrow) partially shown due to U-loop) and left (black arrow) vertebral arteries were still visualised due to adequate intraluminal contrast opacification.