Origin of acute neurovascular syndrome by means of ¹⁸F-sodium fluoride positron emission tomography-magnetic resonance imaging

Abstract

BACKGROUND: Despite the substantial progress made in understanding that high-risk carotid plaque features can more accurately reflect the risk of acute neurovascular events independently of the degree of stenosis, clinicians around the globe continue to rely on the luminal diameter stenosis of carotid artery as the main parameter that stratifies patients to either surgery or best medical therapy. In effect, non-obstructive (<50% stenosis) carotid plaques are excluded from a stroke preventative intervention despite being highly likely to cause thromboembolic complications. Furthermore, although cross-sectional imaging modalities can provide complementary plaque characteristics, neither of currently available imaging systems can offer a robust insight into the biological status of the clinically culprit plaque. Fluorine 18-labelled sodium fluoride positron emission tomography-magnetic resonance imaging (18F-NaF PET-MRI) is the only imaging technique which facilitates parallel assessment of plaque characterisation; cellular (inflammation), molecular (microcalcification) and morphological (macrocalcification, intraplaque haemorrhage, necrotic lipid core, luminal thrombus, and fibrosis). This thesis focused on the use of 18F-NaF PET-MRI to investigate pathobiological processes of the culprit carotid plaques which are most likely to produce neurovascular syndromes.

METHODS AND RESULTS: Transcranial Doppler: Reproducibility Study (Chapter 3) This study evaluated the reproducibility of transcranial Doppler ultrasound in healthy volunteers and patients with symptomatic carotid artery stenosis to explore the role of transcranial Doppler in characterisation of clinically culprit carotid plaque. In healthy volunteers (n=20, 31±9 years, 11 male), within-day repeatability of Doppler measurements was 0.880 (95% CI 0.726–0.950) for peak velocity, 0.867 (95% CI 0.700–0.945) for mean velocity, and 0.887 (95% CI 0.741–0.953) for end-diastolic velocity. Between-day reproducibility was similar but lower: 0.777 (95% CI 0.526–0.905), 0.795 (95% CI 0. 558–0.913), and 0.674 (95% CI 0.349–0.856), respectively. In patients (n=20, 72±11 years, 11 male), within-day repeatability of Doppler measurements was higher: 0.926 (95% CI 0.826–0.970) for peak velocity, 0.922 (95% CI 0.817–0.968) for mean velocity, and 0.868 (95% CI 0.701–0.945) for end-diastolic velocity. Similarly, between-day reproducibility revealed lower values: 0.800 (95% CI 0.567– 0.915), 0.786 (95% CI 0.542–0.909), and 0.778 (95% CI 0.527–0.905), respectively. In both cohorts, the intra-observer Bland Altman analysis demonstrated acceptable mean measurement differences and limits of agreement between series of middle cerebral artery velocity measurements with very few outliers. In patients, the carotid stenoses were 30–40% (n = 9), 40–50% (n = 6), 50–70% (n = 3) and > 70% (n =2). No spontaneous embolisation episodes were detected in either of the groups.

¹⁸F-NaF PET-MRI: Acute Neurovascular Syndrome (Chapter 4) In this prospective observational single-centre cohort study, 110 participants were evaluated (mean age, 68 years 6±10 [SD]; 70 men and 40 women). Of the 110, 34 (32%) had prior cerebrovascular disease, 26 (24%) presented with amaurosis fugax, 54 (49%) with transient ischemic attack, and 30 (27%) with stroke. Compared with non-culprit carotids, culprit carotids had greater stenoses (³50% stenosis: 30% vs 15% [P = 0.02]; ³70% stenosis: 25% vs 4.5% [P, 0.001]) and had increased prevalence of MRI-derived adverse plaque features, including intraplaque haemorrhage (42% vs 23%; P = 0.004), necrotic core (36% vs 18%; P = 0.004), thrombus (7.3% vs 0%; P = 0.01), ulceration (18% vs 3.6%; P = 0.001), and higher 18F-NaF uptake (maximum tissue-to-background ratio, 1.38 [IQR, 1.12–1.82] vs 1.26 [IQR, 0.99–1.66], respectively; P = 0.04). Higher 18F-NaF uptake was positively associated with necrosis, intraplaque haemorrhage, ulceration, and calcification and inversely associated with fibrosis (P = 0.04 to P, 0.001). In multivariable analysis, carotid stenosis at or over 70% (odds ratio, 5.72 [95% CI: 2.2, 18]) and MRI-derived adverse plaque characteristics (odds ratio, 2.16 [95% CI: 1.2, 3.9]) were both associated with the culprit versus non-culprit carotid vessel.

¹⁸F-NaF PET-MRI: Correlation with Histology (Chapter 5) This study was a subset of the prospective observational single-centre cohort study (Chapter 4) as described above that aimed to assess the correlation between ¹⁸F-NaF uptake with ex vivo imaging and histological examination. ¹⁸F-NaF PET-MRI was evaluated using the maximum standard uptake values (SUVₘₐₓ). A total of 20 participants were evaluated (mean age, 69±9 years; 13 male). In vivo 3T MRI demonstrated good to excellent agreement with ex vivo 7T MRI (intraplaque haemorrhage, rupture, or erosion, intraplaque haemorrhage) and histology (lipid core, thrombus, calcium) although agreement was poor for lipid core on ex vivo 7T MRI and rupture or erosion on histology. The level of agreement between ex vivo 7T MRI and histology ranged from good (intraplaque haemorrhage, lipid core) to excellent (calcium), whereas there was a poor agreement for plaque rupture or erosion with no thrombus detection by ex vivo 7T MRI. There was a strong correlation between in vivo and ex vivo ¹⁸F-NaF uptake (SUVₘₐₓ, R=0.78, p<0.001), and they both correlated with calcium volume (in vivo SUVₘₐₓ, R=0.73, P<0.001; ex vivo SUVₘₐₓ, R=0.51, P=0.03) and with calcium index (in vivo SUVmax, R=0.67, P=0.004; ex vivo SUVmax, R=0.62, p=0.008). In vivo ¹⁸F-NaF correlated with elastin index (SUVₘₐₓ, R=0.44, P=0.05), inversely with lipid core (SUVmax, R= -0.55, P=0.02) but not with fibrin index (SUVₘₐₓ, R=-0.17, P=0.48) or red blood cell index (SUVₘₐₓ, R=-0.25, P=0.29). In vivo SUVₘₐₓ of ¹⁸F-NaF uptake predicted histological carotid plaque grade severity (P= 0.04).

CONCLUSIONS Whilst intracranial Doppler is a reproducible technique, spontaneous embolic episodes are rare and do not provide a clear method for risk stratification in patients with stroke. In contrast, in the largest prospective observational data of acute neurovascular syndromes, ¹⁸F-NaF PET-MRI identified clinically culprit carotid plaques that were characterised by adverse plaque features, higher radiotracer uptake and coexistent active aortic vascular lesions. Furthermore, ¹⁸F-NaF PET-MRI uptake co-localised with the most severe histological carotid plaque grade severity providing a comprehensive map of carotid tissue mineralisation supported by relevant ex vivo imaging modalities and histological analysis. At present, ¹⁸F-NaF PET-MRI is the only imaging modality that enables simultaneous structural and physiological atherosclerotic plaque assessment. This may improve further scientific understanding of atherosclerotic disease and potential therapeutic interventions. Most importantly, hybrid PET-MRI as an ultimate imaging platform could transform the management of patients from the current allstenosis- based approach to a truly individualised therapeutic dyad centred on the biological plaque activity.