Author + information
- Published online April 3, 2017.
- Myra S. Cocker, PhD,
- J. David Spence, MD,
- Robert Hammond, MD,
- George Wells, PhD,
- Robert A. deKemp, PhD,
- Cheemun Lum, MD,
- Adebayo Adeeko, PgDip,
- Martin J. Yaffe, PhD,
- Eugene Leung, MD,
- Andrew Hill, MD,
- Sudhir Nagpal, MD,
- Grant Stotts, MD,
- Murad Alturkustani, MD,
- Laurel Hammond,
- Jean DaSilva, PhD,
- Tayebah Hadizad, PhD,
- Jean-Claude Tardif, MD,
- Rob S.B. Beanlands, MD∗ (, )
- Canadian Atherosclerosis Imaging Network (CAIN)
- ↵∗Division of Cardiology, National Cardiac PET Centre, University of Ottawa Heart Institute, 40 Ruskin Street, Ottawa, Ontario, K1Y 4W7, Canada
Although macroscopic calcium deposits in atherosclerotic plaques impart stability, microcalcific deposits can amplify mechanical stress in the fibrous cap by 600 kPa (1). Blood flow, stress, and tension between calcified and noncalcified tissue can increase the risk of plaque rupture. It is postulated that [18F]-sodium fluoride (18F-NaF) imaged with positron emission tomography (PET) replaces the hydroxyl groups of hydroxyapatite, expressed in regions with active microcalcification (2). Validation of 18F-NaF as an imaging-derived biomarker of hydroxyapatite in atherosclerotic plaque is required.
Eleven patients (69 ± 5 years old, 3 women) with high-risk cerebrovascular disease who were scheduled for endarterectomy were recruited. 18F-NaF PET/computed tomography (CT) imaging of the carotid vasculature was acquired within 2 weeks before surgery. Institutional ethics committee review approval was obtained (OHSN-REB 20120224-01H). Patients provided informed consent. Sixty min after 18F-NaF injection (3 MBq/kg), PET/CT imaging was performed, followed by CT angiography (Discovery 690 PET-VCT, GE Medical Systems, Milwaukee, Wisconsin). Radiation effective dose was 9 mSv. One patient could not complete imaging.
18F-NaF PET/CT images were coregistered with a HybridViewer (Hermes Medical Solutions, Greenville, North Carolina). 18F-NaF uptake (Bq/cc) was normalized to injected activity and body weight (standardized uptake value [SUV]). Plaque SUV was normalized to the mean SUV of blood in the internal jugular vein (tissue-to-blood ratio [TBR]). Bilateral carotid 18F-NaF uptake was quantified by: 1) a vessel-based approach to define maximum 18F-NaF uptake for each PET slice spanning 2 cm above and below the bifurcation (TBRvessel, SUVvessel); and 2) a foci-based approach for determining the maximum 18F-NaF activity for each plaque (TBRmax, SUVmax).
En bloc excised plaque was sectioned and stained with Goldner’s trichrome [hydroxyapatite (3)] and Alizarin Red S (extent of calcification). The extent of staining was quantified with color-encoded digitized slides (Aperio Technologies, Vista, California). The carotid bifurcation was identified on digitized 3-dimensional histology images and located on CTA and PET to coregister with histology.
Bilateral carotid plaques were classified as either associated with patient symptoms (transient ischemic attacks or stroke; 9 plaques) or not (11 plaques). Plaque associated with symptoms had evidence for greater 18F-NaF uptake than plaque not associated with symptoms (TBRmax: 3.75 ± 1.10 vs. 2.79 ± 0.60; p = 0.04; SUVmax: 3.00 ± 0.90 vs. 2.40 ± 0.80; p = 0.125). 18F-NaF uptake was related to Goldner’s trichrome expression (TBRvessel: r = 0.45; p < 0.001; SUVvessel: r = 0.43; p < 0.001) but not Alizarin Red S (TBRvessel: r = 0.12; p = 0.36; SUVvessel: r = −0.05; p = 0.73) (Figure 1). To account for intraplaque clustering of data, a fixed-effects model for combining correlations with weights inversely related to within-patient variation was used. This compared the independent and individual correlation between 18F-NaF uptake and staining for each plaque. Findings confirmed the 18F-NaF TBRvessel and Goldner’s trichrome relationship (r = 0.37; p = 0.006) (Figure 1).
Elevated 18F-NaF uptake observed in plaque associated with patient symptoms might reflect active microcalcific processes. Microcalcification could increase the risk of mechanical failure, contributing to plaque instability (1). It is proposed that sodium fluoride replaces the hydroxyl groups of hydroxyapatite (2). Supporting this hypothesis, the present study validates the relationship between in vivo 18F-NaF uptake and ex vivo hydroxyapatite expression within carotid plaque. 18F-NaF uptake was related to active microcalcification (hydroxyapatite expression) rather than the overall extent of calcification (Alizarin Red S), which suggests that 18F-NaF selectively targets the regions of active calcification.
Study limitations include the recognized challenges with coregistration of PET/CT images against histology; the absence of alkaline phosphatase validation, although this has been reported by others (4); and the fact that this was a small proof-of-concept study.
Calcium has been considered to be a static, aplastic amorphous material. This is in part because of the lack of technology available to assess the dynamic nature of calcification, which renders the potential impact and prognostic relevance of active calcification virtually unexplored. Our findings demonstrate that 18F-NaF imaging can noninvasively identify active calcification.
Please note: This work was supported by the Canadian Institutes of Health Research grant #CRI-88057. Dr. Cocker was a Research Fellow of the Heart and Stroke Foundation (HSF) of Canada and is currently the Ottawa Heart Institute Ernest & Margaret Ford Fellow. Dr. Beanlands is a Career Investigator supported by the HSF, Tier 1 Ottawa Research Chair and Vered Chair in Cardiology; and has been a consultant for and received grant funds from GE Healthcare. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose. The findings of this study have been presented in part at the American Heart Association Scientific Sessions 2014 and the Canadian Cardiovascular Congress 2014.
- 2017 American College of Cardiology Foundation