Approval for this prospective study was obtained and informed consent was waived by the ethics committee of our institution, as in our institution a cerebral NCCT, cerebral CTP, and head and neck CTA is routinely performed in the workup of acute ischaemic stroke patients. One or two time points of the CTP acquisition was substituted for a volumetric neck CTA, leaving all other factors unchanged. In a previous retrospective study (part I), we showed that the calculated perfusion values were not significantly affected.
Four senior technicians were trained to use the One-Step Stroke Protocol on our scanner. Consecutive patients with the indication of acute ischaemic stroke were prospectively enrolled and underwent NCCT, the One-Step Stroke Protocol, and conventional head and neck CTA between November 2013 and June 2014. Inclusion criteria were as follows: admission within a 9-h time window from the onset of their symptoms, NIHSS score of at least 2, and patients presenting during office hours while one of the four trained technicians was available for the exam. Exclusion criteria were as follows: patients who did not receive conventional CTA, severe metal artefacts on conventional CTA or the One-Step Stroke Protocol, severe motion artefacts resulting from motion during scanning on conventional CTA or the One-Step Stroke Protocol, known kidney failure and previous allergic reactions to iodinated contrast medium.
CT imaging was performed on a 320-detector row CT scanner (Aquilion ONE; Toshiba Medical Systems, Otawara, Japan). The scan protocol consisted of a cerebral NCCT, the One-Step Stroke Protocol, and conventional head and neck CTA. In all patients two contrast injections were performed, one for the One-Step Stroke Protocol and one for CTA.
NCCT scanning of the brain was performed at 120 kV tube voltage and 280 mA tube current, 1 s rotation time, 0.5 mm section thickness and 0.5 mm reconstruction interval.
For the One-Step Stroke Protocol, 50 ml non-ionic contrast agent (300 mg iodine/ml Xenetix 300; Guerbet, Villepinte, France) was injected into an antecubital vein with an injection rate of 5 ml/s, followed by a 40-ml saline flush at 5 ml/s. Whole-brain volumetric acquisitions (16 cm z-coverage) were acquired with 0.5 mm slice thickness, 0.5 s rotation time and 80 kV tube voltage. The CTP acquisition started 5 s after contrast injection, with 1 volumetric acquisition at 200 mAs, followed after 4 s by 12 volumetric acquisitions each at 100 mAs with 2 s sampling interval, followed after 5 s by five acquisitions each at 75 mAs with 5 s interval. The total number of volumetric acquisitions was 18 and the total scan duration was 60 s (Fig. 1). Image reconstruction was done with reconstruction kernel FC41 and standard AIDR3D (adaptive iterative dose reduction in three dimensions; Toshiba Medical Systems).
During the CTP acquisition the table was moved to the neck and back to the brain to perform a neck CTA. This movement required 2 × 1.8 s with our CT scanner. The neck CTA can be acquired in 0.275–0.500 s. Therefore, the time gap induced between adjacent CTP acquisitions can theoretically be less than 4 s. At present, the neck CTA is initiated manually as soon as arterial contrast enhancement is visible on the mid-section of the brain, which is presented as a real-time reconstruction on the scanner console. This manual interaction increases the time gap slightly. Given a scanning sequence in which data are acquired every 2 s, the time gap needed to acquire the neck CTA is in the range of one to two volumetric acquisitions of the CTP (Fig. 1).
Since bolus tracking during CTP is not available yet, a manual interaction of the technician was needed to start the neck CTA as soon as contrast material was visible in the central slice of the brain during the CTP acquisition. The table then moved to the neck within 1.8 s. The volumetric neck CTA (with 16 cm z-coverage) was acquired with following parameters: 0.5 mm slice thickness, 0.5 s rotation time, 80 kV tube voltage and 200 mAs exposure. FC43 filter was used for image reconstruction. After acquiring the neck CTA, the table moved back to the brain to resume the CTP acquisition.
For the head and neck CTA, 70-80 ml non-ionic contrast agent (300 mg iodine/ml Xenetix 300; Guerbet, Villepinte, France) was injected into the antecubital vein with an injection rate of 5 ml/s followed by a 40-ml saline flush with an injection rate of 5 ml/s. The CTA covered the area from just below the aortic arch to the vertex. CTA was performed in a helical scan mode using the following parameters: 80 × 0.5 collimation, 0.81 pitch, 120 kV, automatic exposure control with standard deviation of 10 and exposure range 100-700 mA, 0.5 mm and 3.0 mm slice thickness, 0.5 s rotation time, reconstruction filter FC43 and standard AIDR3D. The bolus tracker was set at an absolute threshold of 180 HU at the level of the descending aorta.
The distance from the lowest origin of the common carotids to the vertex was measured on conventional CTA to determine the ideal total z-coverage. Given a margin of 1 cm for the coverage of the One-Step Stroke Protocol, we determined how many patients could have had complete coverage of the craniocervical circulation covered under ideal planning of the scan range. For this we determined the number of patients in whom this distance was less than 31 cm. The distance missed of the origin of the left common carotid artery by the One-Step Stroke Protocol was recorded in centimetres. Also, the heights of the patients were reported.
In addition, we measured the time interval between the last CTP acquisition, before performing the volumetric neck CTA, to the first CTP acquisition thereafter.
Arterial enhancement of the carotid and vertebral arteries was measured at three levels: just above the origin of the common carotid artery (CCA), in the internal carotid artery (ICA) above the bifurcation and at the level of the C1-C2. Enhancement of the vertebral arteries was measured at the same three levels. For contrast-to-noise ratio (CNR) the sternocleidomastoid muscle and the noise represented by the standard deviation of the HU values in a region of interest (ROI) in the surrounding air was used. The ROIs were kept constant at 4 mm2 and 64 mm2 for the arterial enhancement and sternocleidomastoid muscle, respectively. Calcifications and plaques were avoided.
Three observers (F.J.A.M., E.J.S. and M.P. with 10, 5 and 20 years of experience in stroke imaging, respectively) qualitatively scored image quality at several levels of the neck CTA acquisitions. Data was anonymised and observers were blinded to the technique. This was realised by cropping both images of the conventional CTA and the One-Step Stroke Protocol neck CTA such that only the overlapping field of view of both techniques was visible and that the z-coverage could not reveal the used scanning technique. Image quality was scored on a scientific scoring workstation Cirrus, developed at the Diagnostic Image Analysis Group (DIAG), Nijmegen, The Netherlands.
Image quality of the carotid and vertebral artery was determined on the following five-point scale: 1, non-diagnostic; 2, poor image quality, sufficient for vascular evaluation; 3, moderate image quality, sufficient for soft tissue and vascular evaluation; 4, good image quality; 5, excellent image quality .
Streak and pulsation artefacts were scored as follows: 1, severe artefacts, non-diagnostic quality; 2, substantial artefacts, moderate impairment of diagnostic quality; 3, clearly visible artefacts, no impairment of diagnostic quality; 4, hardly visible artefacts; 5, no artefacts.
Artefacts and image quality of cervical arteries were scored at the same three anatomic levels that were used for quantitative evaluation of the arterial enhancement.
An expert observer (F.J.A.M.) assessed vCTA and conventional CTA for the presence of stenoses, occlusions, dissections, and coverage. Stenoses were graded as significant (more than 50 % luminal stenosis) or non-significant (less than 50 % luminal stenosis). The observer was blinded to technique, clinical information and diagnoses. The images were shown in a random order. Multiplanar reconstructions (MPRs) were available. Scans were presented with equal display settings, but the observer could change the settings according to clinical practice (window levelling, arbitrary planes, slab thickness).
Statistical analyses were performed using the Statistical Package of Social Sciences version 20.0 for Windows (SPSS, Chicago, USA). Wilcoxon signed rank test was used to test for significant differences between the image quality of the conventional CTA and the volumetric CTA from the One-Step Stroke Protocol. A P value of <0.05 was considered significant.