Imaging of intracranial arterial disease: a comparison between MRI and unenhanced CT.

Background and purpose: Arterial calcifications on unenhanced CT scans and vessel wall lesions on MRI are often used interchangeably to portray intracranial arterial disease. However, the extent of pathology depicted with each technique is unclear. We investigated the presence and distribution of these two imaging findings in patients with a history of cerebrovascular disease. Materials and methods: We analyzed CT and MRI data from 78 patients admitted for stroke or TIA at our institution. Vessel wall lesions were assessed on 7 T MRI sequences, while arterial calcifications were assessed on CT scans. The number of vessel wall lesions, severity of intracranial internal carotid artery (iICA) calcifications, and overall presence and distribution of the two imaging findings were visually assessed in the intracranial arteries. Results: At least one vessel wall lesion or arterial calcification was assessed in 69 (88%) patients. Only the iICA and vertebral arteries (VA) showed a substantial number of both calcifications and vessel wall lesions. The other vessels showed almost exclusively vessel wall lesions. The number of vessel wall lesions was associated with the severity of iICA calcification (p = 0.013). Conclusions: The number of vessel wall lesions increases with the severity of iICA calcifications. Nonetheless, the distribution of vessel wall lesions on MRI and arterial calcifications on CT shows remarkable differences. These findings support the need for a combined approach to examine intracranial arterial disease.

RISK FACTORS FOR MEDIAL AND INTIMAL INTRACRANIAL INTERNAL CAROTID ARTERY CALCIFICATION IN MEN AND WOMEN WITH CARDIOVASCULAR DISEASE. THE UCC-SMART STUDY.

INTRODUCTION: Calcifications of the intracranial internal carotid artery (iICA) can lead to an increased risk for stroke. Two types of iICA calcification are known: those affecting the tunica intima or the tunica media. In extracranial arteries, risk factors and calcification patterns are different in women and men, but little is known regarding the iICA. In this study we aimed to identify sex-specific risk profiles and medications associated to intimal and medial iICA calcification in patients with cardiovascular disease (CVD). METHODS: Participants of the UCC-SMART cohort undergoing a non-contrast head CT within six months from the study inclusion were considered (n=475). Intimal or medial iICA calcification pattern was assessed using a previously histology-validated method. Sex-stratified associations between calcification pattern and cardiovascular risk factors, laboratory parameters, and medication use were calculated using Poisson regression analysis with robust standard errors. RESULTS: 204 women and 271 men (age range 24-79 years) were included. 45.4% of men and 34.8% of women showed intimal iICA calcification, while 28.4% of men and 24.0% of women showed medial iICA calcification. Minimal or no iICA calcification was observed in 26.2% of men and in 41,2% of women (reference group). Older age was associated with both calcification patterns in women and men. In women, use of vitamin K antagonists and lipid lowering drugs were associated to medial calcification, while systolic blood pressure and glucose levels were associated to intimal calcification. In men, current smoking was associated to intimal calcification. CONCLUSIONS: Women and men with CVD show differences in risk profiles and medication use associated to intimal and medial iCA calcification.

The association between antiplatelet therapy and changes in intraplaque hemorrhage in patients with mild to moderate symptomatic carotid stenosis: a longitudinal MRI study.

INTRODUCTION: Carotid atherosclerotic intraplaque hemorrhage (IPH) predicts stroke. Patients with a history of stroke are treated with antiplatelet agents to prevent secondary cardiovascular events. A positive association between previous antiplatelet use and IPH was reported in a cross-sectional analysis. We investigated changes in IPH over two years in patients who recently started versus those with continued antiplatelet use. METHODS: In the Plaque at Risk (PARISK) study, symptomatic patients with <70% ipsilateral carotid stenosis underwent carotid plaque MRI at baseline and after two years to determine IPH presence and volume. Participants were categorized into new users (starting antiplatelet therapy following the index event) and continued users (previous use of antiplatelet therapy before the index event). The association between previous antiplatelet therapy and the presence of IPH at baseline MRI was investigated using multivariable logistic regression analysis. IPH volume change over a period of two years, defined as the difference in volume between follow-up and baseline, was investigated in each group with a Wilcoxon signed-rank test. The IPH volume change was categorized as progression, regression, or no change. Using multivariable logistic regression, we investigated the association between new antiplatelet use and 1) newly developed ipsilateral or contralateral IPH and 2) IPH volume progression. RESULTS: A total of 108 patients underwent carotid MRI at baseline and follow-up. At baseline, previous antiplatelet therapy was associated with any IPH (OR=5.6, 95% CI: 1.3-23.1; p=0.02). Ipsilateral IPH volume did not change significantly during the two years in patients who continued receiving antiplatelet agents (86.4 mm3 [18.2-235.9] vs. 59.3 mm3 [11.4-260.3]; p=0.6) nor in the new antiplatelet users (n=31) (61.5 mm3 [0.0-166.9] vs. 27.7 mm3 [9.5-106.4]; p=0.4). Similar results of a nonsignificant change in contralateral IPH volume during those two years were observed in both groups (p>0.05). No significant associations were found between new antiplatelet use and newly developed IPH at two years (odds ratio (OR)=1.0, 95% CI:0.1-7.4) or the progression of IPH (ipsilateral: OR=2.4, 95% CI:0.3-19.1; contralateral: OR=0.3, 95% CI:0.01-8.5). CONCLUSION: Although the baseline association between IPH and previous antiplatelet therapy was confirmed in this larger cohort, the new onset of antiplatelet therapy after TIA/stroke was not associated with newly developed IPH or progression of IPH volume over the subsequent two years.

Ischemic stroke recurrence and mortality in different imaging phenotypes of ischemic cerebrovascular disease: The SMART-MR Study.

BACKGROUND: Diagnosis of cerebrovascular disease is based on both clinical and radiological findings, however, they do not always correlate. AIMS: To investigate ischemic stroke recurrence and mortality in patients with different imaging phenotypes of ischemic cerebrovascular disease. METHODS: Within the SMART-MR study, a prospective patient cohort with arterial disease, cerebrovascular diseases of participants at baseline were classified as no cerebrovascular disease (reference group, n = 828), symptomatic cerebrovascular disease (n = 204), covert vascular lesions (n = 156), or imaging negative ischemia (n = 90) based upon clinical and MRI findings. Ischemic strokes and deaths were collected at 6 month-intervals up to 17 years of follow-up. With Cox regression, relationships between phenotype and ischemic stroke recurrence, cardiovascular mortality, and non-vascular mortality were studied adjusted for age, sex, and cardiovascular risk factors. RESULTS: Compared to reference group risk for recurrent ischemic stroke was increased not only in the symptomatic cerebrovascular disease (HR 3.9, 95% CI 2.3-6.6), but also in the covert vascular lesion (HR 2.5, 95% CI 1.3-4.8) and the imaging negative ischemia groups (HR 2.4, 95% CI 1.1-5.5). Risk for cardiovascular mortality was increased in the symptomatic cerebrovascular disease (HR 2.2, 95% CI 1.5-3.2) and covert vascular lesions groups (HR 2.3, 95% CI 1.5-3.4), while the risk was less strong but also increased in the imaging negative ischemia group (HR 1.7, 95% CI 0.9-3.0). CONCLUSIONS: People with all imaging phenotypes of cerebrovascular disease have increased risk of recurrent ischemic stroke and mortality compared to other arterial diseases. Strict preventive measures should be performed even when imaging findings or clinical symptoms are absent. DATA ACCESS STATEMENT: For use of anonymized data, a reasonable request has to be made in writing to the UCC-SMART study group and the third party has to sign a confidentiality agreement.

Intracranial atherosclerosis in pseudoxanthoma elasticum: A case-control study.

BACKGROUND AND AIMS: Pseudoxanthoma elasticum (PXE) is a genetic disorder characterized by systemic calcification of elastin fibers. Additionally, PXE is associated with an increased risk of stroke. It has been hypothesized that this may be caused by accelerated (intracranial) atherogenesis, as a consequence of specific genetic mutations underlying PXE. Hence, we compared the distribution and burden of intracranial atherosclerosis between PXE patients and healthy controls. METHODS: Fifty PXE patients and 40 age-and-sex-matched healthy controls (without previous ischemic cerebrovascular disease) underwent 3T MRI to visualize atherosclerotic intracranial vessel wall lesions (VWLs). We compared the presence and burden of VWLs (total and for the anterior cerebral, middle cerebral, intracranial internal carotid, posterior cerebral, and basilar arteries separately) between PXE patients and healthy controls using logistic (presence versus absence) and negative binomial regression models (VWL count) adjusted for relevant confounders. All regressions included group (PXE patients vs. healthy controls) as independent variable. RESULTS: We found that 34 (68.0%) PXE patients and 28 (70.0%) healthy controls had a VWL (odds ratio for presence 1.06 [95%CI 0.38-2.91]). In addition, the total burden of VWLs was similar between PXE patients (68 VWLs) and healthy controls (73 VWLs, incidence rate ratio for count 0.81 [95%CI 0.55-1.20]). Findings were similar when analyses were stratified for artery. CONCLUSIONS: The distribution and burden of intracranial atherosclerosis were similar between PXE patients and healthy controls. This implies PXE and its underlying mutations do not involve increased (intracranial) atherogenesis and that vascular calcification or other mechanisms explains the increased stroke risk in PXE.

Lipoprotein(a) levels and atherosclerotic plaque characteristics in the carotid artery: The Plaque at RISK (PARISK) study.

BACKGROUND AND AIMS: Lipoprotein(a) is an independent risk factor for cardiovascular disease and recurrent ischemic stroke. Lipoprotein(a) levels are known to be associated with carotid artery stenosis, but the relation of lipoprotein(a) levels to carotid atherosclerotic plaque composition and morphology is less known. We hypothesize that higher lipoprotein(a) levels and lipoprotein(a)-related SNPs are associated with a more vulnerable carotid plaque and that this effect is sex-specific. METHODS: In 182 patients of the Plaque At RISK study we determined lipoprotein(a) concentrations, apo(a) KIV-2 repeats and LPA SNPs. Imaging characteristics of carotid atherosclerosis were determined by MDCTA (n = 161) and/or MRI (n = 171). Regressions analyses were used to investigate sex-stratified associations between lipoprotein(a) levels, apo(a) KIV-2 repeats, and LPA SNPs and imaging characteristics. RESULTS: Lipoprotein(a) was associated with presence of lipid-rich necrotic core (LRNC) (aOR = 1.07, 95% CI: 1.00; 1.15), thin-or-ruptured fibrous cap (TRFC) (aOR = 1.07, 95% CI: 1.01; 1.14), and degree of stenosis (ß = 0.44, 95% CI: 0.00; 0.88). In women, lipoprotein(a) was associated with presence of intraplaque hemorrhage (IPH) (aOR = 1.25, 95% CI: 1.06; 1.61). In men, lipoprotein(a) was associated with degree of stenosis (ß = 0.58, 95% CI: 0.04; 1.12). Rs10455872 was significantly associated with increased calcification volume (ß = 1.07, 95% CI: 0.25; 1.89) and absence of plaque ulceration (aOR = 0.25, 95% CI: 0.04; 0.93). T3888P was associated with absence of LRNC (aOR = 0.36, 95% CI: 0.16; 0.78) and smaller maximum vessel wall area (ß = -10.24, 95%CI: -19.03; -1.44). CONCLUSIONS: In patients with symptomatic carotid artery stenosis, increased lipoprotein(a) levels were associated with degree of stenosis, and IPH, LRNC, and TRFC, known as vulnerable plaque characteristics, in the carotid artery. T3888P was associated with lower LRNC prevalence and smaller maximum vessel wall area. Further research in larger study populations is needed to confirm these results.

Association of Ischemic Imaging Phenotype With Progression of Brain Atrophy and Cerebrovascular Lesions on MRI: The SMART-MR Study.

BACKGROUND AND OBJECTIVE: To investigate the association of silent vascular lesions, imaging negative ischemia, and symptomatic cerebrovascular disease with long-term progression of brain atrophy and cerebrovascular lesions in patients with arterial disease. METHODS: Within the Second Manifestations of Arterial Disease-Magnetic Resonance (SMART-MR) study, stroke status of participants at baseline was classified as no cerebrovascular disease (reference group, n = 829), symptomatic cerebrovascular disease (n = 206), silent vascular lesion (n = 157), and imaging-negative ischemia (n = 90) according to clinical and MRI findings. With the use of linear mixed models, changes in brain and white matter hyperintensity (WMH) volumes at baseline and during 12 years of follow-up were studied in stroke classifications. Relative risks were estimated for new infarcts during follow-up associated with stroke classifications. Analyses were adjusted for age, sex, cardiovascular risk factors, and medications. RESULTS: Symptomatic cerebrovascular disease associated with 0.35 SD (95% confidence interval [CI] 0.24-0.47) smaller brain volume and 0.61 SD (95% CI 0.48-0.74) larger WMH volume at baseline and increased risk for new infarcts during follow-up (risk ratio [RR] 2.89, 95% CI 2.00-4.16). Silent vascular lesions were associated with 0.15 SD (95% CI 0.01-0.88) smaller brain volume, 0.02 SD (95% CI 0.01-0.03) steeper brain atrophy slope, and 0.48 SD (95% CI 0.32-0.64) larger WMH volume at baseline, in addition to increased risk for lacunes (RR 2.08, 95% CI 1.48-2.94). Individuals with imaging-negative ischemia had increased risk for cortical infarcts (RR 2.88, 95% CI 2.17-3.82). DISCUSSION: Patients with symptomatic cerebrovascular disease, silent vascular lesions, or imaging-negative ischemia have a different course of brain volume loss and cerebrovascular lesion development. These findings may have implications for future stroke risk and dementia and need further investigation.

Risk factors for calcification of the vertebrobasilar arteries in cardiovascular patients referred for a head CT, the SMART study.

BACKGROUND AND PURPOSE: Vertebrobasilar artery calcification (VBAC) has been associated with increased stroke occurrence. Little is known on VBAC risk factors, especially for patients with cardiovascular disease. We aimed to assess risk factors associated with VBAC in a cohort of cardiovascular patients referred for a head computed tomography (CT) scan. MATERIALS AND METHODS: All patients who underwent a clinically indicated, unenhanced, thin slice head CT 6 months before or after inclusion in the SMART study were included. CTs were assessed for presence of VBAC (dichotomously). Relative risks of the associations of age, sex, diabetes mellitus (DM), obesity, body mass index, estimated glomerular filtration rate, hypertension, hyperlipidemia, use of lipid lowering medication, smoking status, high sensitivity C-reactive protein, ankle-brachial index (ABI; ≤0.90, ≥1.30, continuous), internal carotid artery stenosis ≥70%, and carotid intima-media thickness (IMT) with VBAC were estimated using Poisson regression analysis with robust standard errors, adjusted for age and sex. RESULTS: Of the 471 patients included (57% male, median age 58 [interquartile range 47-63]), 117 (24.8%) showed VBAC. Presence of VBAC was associated with older age (RR per 10 years=1.70 [95%CI 1.46-1.99]), DM (RR=1.45 [95%CI 1.03-2.06]), obesity (RR=1.53 [95%CI 1.10-2.12]), ABI ≤0.90 (RR=1.57 [95%CI 1.02-2.41]), and an increased carotid IMT (RR=2.60 per mm [95%CI 1.20-5.62]). Other measurements were not associated with VBAC. CONCLUSIONS: We identified several markers associated with VBAC in patients with cardiovascular disease referred for a head CT. Future investigation into the relationship between VBAC and stroke is warranted to determine the potential of VBAC in stroke prevention.

Microstructural White Matter Abnormalities and Cognitive Impairment After Aneurysmal Subarachnoid Hemorrhage.

Background and Purpose- Aneurysmal subarachnoid hemorrhage (aSAH) may have detrimental effects on white matter microstructure, which may in turn explain the cognitive impairments that occur often after aSAH. We investigated (1) whether the white matter microstructure is altered in patients with aSAH compared with patients with an unruptured intracranial aneurysm and (2) whether these abnormalities are associated with cognitive impairment 3 months after ictus. Methods- Forty-nine patients with aSAH and 22 patients with an unruptured intracranial aneurysm underwent 3T brain magnetic resonance imaging, including a high-resolution diffusion tensor imaging sequence. Patients with aSAH were scanned 2 weeks and 6 months after ictus. Microstructural white matter alterations were quantified by the fractional anisotropy and mean diffusivity (MD). Cognition was evaluated 3 months after ictus. Results- Patients with aSAH had higher white matter MD 2 weeks after ictus than patients with an unruptured intracranial aneurysm (mean difference±SEM, 0.3±0.01×10-3 mm2/s; P≤0.01), reflecting an abnormal microstructure. After 6 months, the MD had returned to the level of the unruptured intracranial aneurysm group. No between-group differences in fractional anisotropy were found (-0.01±0.01; P=0.16). Higher MD at 2 weeks was associated with cognitive impairment after 3 months (odds ratio per SD increase in MD, 2.6; 95% CI, 1.1-6.7). The association between MD and cognitive impairment was independent of conventional imaging markers of aSAH-related brain injury (ie, cerebral infarction, hydrocephalus, total amount of subarachnoid blood, total brain volume, or white matter hyperintensity severity). Conclusions- Patients with aSAH have temporary white matter abnormalities in the subacute phase that are associated with cognitive impairment at 3 months after ictus.