Keywords:
Ischaemia / Infarction, Calcifications / Calculi, Arteriosclerosis, Diagnostic procedure, Contrast agent-intravenous, Computer Applications-General, CT-Quantitative, CT-Angiography, CT, Vascular, CNS, Arteries / Aorta
Authors:
V. Rafailidis, I. Chryssogonidis, C. Xerras, T. Tegos, I. NIKOLAOU, A. Charitanti-Kouridou, E. Destanis, A. Kalogera-Fountzila; Thessaloniki/GR
DOI:
10.26044/ecr2019/C-0161
Methods and materials
Patient population
The present study was observational and the patients studied were recruited on a prospective and consecutive pattern from the Radiology and Neurology Department of the Hospital.
Institutional ethics review board approval and written informed consent from every patient were obtained.
Patients who were considered candidates for inclusion in the study were all those referred for carotid CTA after being diagnosed with carotid atherosclerotic disease on US,
either being symptomatic or asymptomatic based on clinical examination and brain imaging.
The cause of referral could either be the occurrence of transient ischemic attack or stroke (symptomatic patients) or other unrelated conditions (such as pre-operative work-up in asymptomatic patients).
Both carotid systems of every patient were included in the study and analyzed separately.
A plaque was considered symptomatic if ipsilateral to a stroke occurring during the last six-month period or asymptomatic if located on the contralateral side or ipsilateral to the side of stroke but after the six-month period.
For a plaque to be considered symptomatic,
a stroke needed to be diagnosed on brain imaging with magnetic resonance imaging (MRI).
The primary inclusion criterion was the identification of an internal carotid artery plaque with moderate (50-69%) or severe (70-99%) stenosis as CTA is only performed if such a plaque is detected on US,
based on the Radiology Department’s standard of care.
Exclusion criteria used in this study included any contra-indication to the use of CTA contrast agents like history of allergy as well as presence of other diseases mimicking stroke or comorbidities that could cause stroke (including arrhythmias,
cardiac anatomic abnormalities,
thrombophilia and immunologic diseases like antiphospholipid syndrome).
These comorbidities were excluded in an attempt to accurately associate the stroke with carotid disease.
If a patient fulfilled the inclusion criteria and had no exclusion criteria,
his CTA examination was exported in digital imaging and communications in medicine (DICOM) format and was used for quantitative analysis with the specialized software in a computer.
Imaging technique
CTA examinations were performed with a 128-slice multi-detector CT system (GE Optima CT660,
GE Healthcare),
both with an unenhanced scan and an angiographic scan.
The scan range covered the area from the ascending aorta and up to the intracranial arteries,
at the level of the frontal sinuses.
Eighty milliliters of contrast agent (37% iodine,
iopromide,
Ultravist,
Bayer) were intravenously administered in a bolus followed by 50 ml of saline bolus chaser,
both administered at an injection rate of 4 ml/s.
Real-time bolus tracking was performed at the level of the ascending aorta and used in order to synchronize contrast passage with the angiographic data acquisition.
The slice thickness of the images reviewed was 0.625 mm for optimal isotropic imaging.
Image analysis
Commercially available imaging analysis software (vascuCAPTM,
Elucid Bioimaging) has been used for the quantification of carotid plaque components.
One board-certified radiologist with 10 years of experience on carotid CTA was specifically trained for the use of this software and performed the quantitative analysis.
The reader initially defined the common,
internal and external carotid artery.
A center-line was then computed by the software and lumen and outer wall segmentation were performed.
Blooming artifact,
blurring of the image and partial volume effects are well-established factors limiting the accuracy of CTA measurements of plaque density.
The process is further complicated by the fact that density may vary even within areas of the same tissue type,
also depending on contrast agent uptake.
The software used in this study makes use of previously described and tested algorithms which are histology-validated and mitigate known limitations of blurring,
calcium blooming,
partial volume effects of routine CTA acquisitions and overly strict dependence on HU thresholds.
These algorithms have been previously described in detail [9,
10].
After computation of vessel center-line and segmentation of lumen and wall,
the software determines the scanner blur based on the lumen boundary and can thus optimize component densities evaluated at subvoxel boundaries in an attempt to best fit the presented image.
After initial segmentation of wall and plaque components,
the reader had the opportunity to manually edit the automated results using the software’s interactive tools,
if necessary.
The plaque components whose volume was quantified using this software included: lipid core,
fibrous matrix and calcification.
Both absolute and relative volumes were calculated,
with the latter being calculated by dividing absolute values by the sum of all absolute values.
The ratio of lipid and calcification (plus one) volume (L/C) was also calculated.
The degree of diametric internal carotid artery stenosis was manually measured using the European Carotid Surgery Trial (ECST) method.
Statistical analysis
The IBM SPSS Statistics version 23.0 was used for statistical analysis.
Descriptive statistics included mean and standard deviation (SD) for normally distributed variables and median and inter-quartile range (IQR) for non-normally distributed variables.
The Kolmogorov-Smirnov test was used to test the normal distribution of variables.
Mann-Whitney U test and t-test were used to compare means between groups depending on the normality of distribution.
Receiver Operating Characteristic (ROC) analysis was used for diagnostic accuracy analysis for the detection of symptomatic plaque.
Statistical significance level was set at 0.05.