Dr. Valentin Fuster,
Dr. James HF. Rudd
Background: Fluorodeoxyglucose positron emission tomography (FDG PET) imaging of
atherosclerosis has been used to quantify plaque inflammation and to measure the effect of plaque stabilizing drugs. Here we explore how atherosclerotic plaque inflammation varies across arterial territories and how it relates to arterial calcification. We also test the hypotheses that the degree of local arterial inflammation measured by PET is correlated with the extent of systemic inflammation and presence of risk factors for vascular disease.
Methods and Results: Forty-one subjects underwent vascular PET/CT imaging with FDG. All had either vascular disease or multiple risk factors for it. Forty subjects underwent carotid imaging, twenty-seven underwent aortic, twenty-four iliac and thirteen femoral imaging. Thirtythree subjects had a panel of biomarkers analyzed. We found strong associations between FDG uptake in neighboring arteries (left vs. right carotid r=0.91, p<0.001, ascending aorta vs. aortic arch r=0.88, p<0.001). Calcification and inflammation rarely overlapped within arteries – carotid artery FDG uptake vs. calcium score r=-0.42, p=0.03). Carotid artery FDG uptake was greater in those with a history of coronary artery disease (target to background ratio (TBR) 1.83 vs. 1.61, p<0.01), and in males vs. females (TBR 1.83 vs. 1.63, p<0.05). Similar findings were also noted in the aorta and iliac arteries. Subjects with the highest levels of FDG uptake also had the greatest concentrations of inflammatory biomarkers: descending aorta TBR vs. matrix metalloproteinase 3 (MMP 3): r=0.53, p=0.01 and carotid TBR vs. MMP 9: r=0.50, p=0.01. Non-significant positive trends were seen between FDG uptake and levels of interleukin 18, fibrinogen and C-reactive protein. Finally, we found that the atheroprotective biomarker adiponectin was negatively correlated with the degree of arterial inflammation in the descending aorta: r=-0.49, p=0.03).
Q: Dr. Rudd, it is great to have you as the leading author on the paper entitled “The Relationships Between Regional Arterial Inflammation, Calcification, Risks Factors and Biomarkers – A Prospective FDG PET/CT Imaging Study”, recently published in Circulation Cardiovascular Imaging. It is exciting to read through it, certainly a comprehensive study. What do you consider as the key findings of the study?
A: Many thanks for the opportunity to discuss the paper. Firstly, I must say that the work was a highly collaborative effort by several researchers in Professor Zahi Fayad's lab at Mount Sinai in New York, including radiologists, cardiologists and imaging technologists. It was funded by the NIH, British Heart Foundation and the Cambridge Biomedical Research Centre. My current lab in Cambridge, UK continues to collaborate with Prof Fayad on several ongoing studies.
There are several interesting findings to come out of the study. First is the fact that the degree of inflammation seems to correlate well between neighboring arteries, for example between left and right carotid arteries. The further the arteries are from each other in the body, so the degree of correlation decreases. Secondly, being male and having a history of past CAD makes arterial inflammation more likely. Finally, as might be expected, arterial inflammation levels were associated with the levels of several circulating inflammatory biomarkers.
Q: Indeed, important findings. What about the observation of rare overlap of plaque inflammation and calcification. Can we conclude that atherosclerotic plaques are either “hot” or “cold” or is a there an in-between?
A: This study unfortunately does not allow us to answer that question fully because we only performed imaging at a single time-point. However, we did find a consistent negative correlation between calcification and inflammation. Other studies, for example, that performed by Ben-Haim, that imaged over several years did show a gradual change in plaque from highly inflamed to calcified. So I don't think that we can say that plaques are either hot of cold, there seems to be a continuum.
Q: The view has been presented that calcification represents a late, burnt-out stage of atherosclerosis which sounds inert. What is therefore significance of this view for coronary calcium imaging, which has been suggested to be used to re-stratify cardiac risk? Should we rather use inflammation imaging?
A: I think that the risk associated with having a high calcium score in the coronary arteries derives predominantly from the fact that patients with many calcified plaques are the likeliest to have many non-calcified plaques as well. Calcium scoring is a robust, reproducible technique with many years of outcome data linking calcium scores with cardiac events. Although, I personally think that a measure of inflammation in the coronary arteries would be very helpful, it is too early to suggest using PET for this purpose. Especially because we would have to contend with cardiac and respiratory motion. Finally, FDG would not be the tracer of choice because it is avidly taken up by the myocardium. We would need more specific macrophage markers for this application.
Q: Your study supports the concept of atherosclerosis as a systemic disease and that the biological activity in one territory reflects biological activity elsewhere in the vasculature, at least the peripheral vasculature in your study even though regional variation was noted. Would you argue that we can extrapolate the findings to the coronary artery circulation? Is there a peripheral territory that correlates the best with the coronary circulation?
A: It is not possible to answer that question based on the results of this study alone. But, my colleague and collaborator Dr Ahmed Tawakol at Massachusetts General Hospital has attempted coronary artery imaging with FDG PET. He has noted that the FDG signal from the ascending aorta frequently correlates with that measured in the coronary arteries.
Q: Intriguingly, the current study shows a correlation with circulating levels of MMP-3 and MMP-9. What about CRP and hs-CRP, particularly in view of recent clinical trials?
A: I am not an expert in biomarkers, but we chose the panel of markers to measure based on several aspects of plaque biology. These included inflammation, thrombogenicity and matrix turnover. Whilst we did see relationships between arterial FDG uptake and several of the MMP family, we also noted a non-significant trend between arterial inflammation and CRP levels. I think our sample size was too small to make any conclusion about which biomarkers were most predictive of inflammation, but the findings were intriguing, and generally supportive of the fact that FDG uptake reflects active arterial inflammation.
Q: Conceivably, some risk factors might be more closely related to plaque activity than other, for instance, LDL and HDL. Did you observe this in your study?
A: We did not see any relationship between FDG uptake and serum lipid levels. But male sex and history of CAD were predictive of arterial inflammation.
Q: In your opinion, should we favor plaque inflammation imaging or circulating biomarkers to assess atherosclerotic disease activity and prognosis? Which imaging technique is most promising? Will imaging be more cost effective one day? What word would you give to clinicians?
A: Right now, I would say what is most needed are outcome studies that link imaging findings (CT, MRI, FDG PET), biomarkers, risk factors and clinical events. The HRP study below is ideally placed to help us answer your question. Until that time, imaging measures of arterial inflammation, such as FDG PET, are probably most useful for the assessment of efficacy of novel anti-inflammatory drugs. Several such studies are currently underway.
Q: You mention the HRP initiative. What does this initiative include?
A: The HRP initiative (High-Risk Plaque - http://www.bg-medicine.com/content/initiatives/hrp).
This initiative is well explained on the website above. Essentially it recruits a group of over 7000 asymptomatic subjects with risk factors for cardiovascular disease and performs baseline clinical assessment including ankle-brachial pressure measurements, arterial imaging (carotid IMT, cardiac CT with calcium, MRI plaque and arterial FDG PET) and biomarkers on them. The subjects are then followed for 3 years and cardiovascular events are collected. The study is unique in that it uses mobile scanners at several US sites to gather the data, reducing variability. It also uses insurance claim and prescription data, obviously with informed consent, to gather information about events. This very exciting study is due to report in 2011. It will help us understand the links between imaging and hard CV endpoints.
Q: What can we do in Europe? Shall we start such initiatives in Europe as well? What alternative steps can we take as a research community?
A: I think it would be very useful to have such initiatives running in Europe. I would imagine many centres would be capable of performing this type of vascular imaging. I believe networks of hospitals well-trained in imaging would be useful for multi-centre studies of both drug therapies and prospective studies of the HRP type. We are seeking already to put such consortia together.
Q: Thank you so much, Dr. Rudd. We really appreciate your insight and time.
A: No problem. I would be happy to address any further questions.
This study shows that FDG PET imaging can increase our knowledge of how atherosclerotic plaque inflammation relates to calcification, serum biomarkers and vascular risk factors. Plaque inflammation and calcification rarely overlap, supporting the theory that calcification represents a late, burnt-out stage of atherosclerosis. Inflammation in one arterial territory is associated with inflammation elsewhere, and the degree of local arterial inflammation is reflected in the blood levels of several circulating biomarkers. We suggest that FDG PET imaging could be used as a surrogate marker of both atherosclerotic disease activity and drug effectiveness. Prospective, event driven studies are now underway to determine the role of this technique in clinical risk prediction.
Circ Cardiovasc Imaging published online Jan 26, 2009; DOI:10.1161/CIRCIMAGING.108.811752
For further questions, please contact Dr. James Rudd at firstname.lastname@example.org
European Society of Cardiology
European Heart HouseLes Templiers2035 Route des CollesCS 80179 Biot
06903Sophia Antipolis, FR
Our mission: To reduce the burden of cardiovascular disease
© 2018 European Society of Cardiology. All rights reserved