Coronary Plaque and the Adjacent Fat: Monitoring Change Over Time∗
Computed tomography coronary angiography (CTCA) now allows for the rapid, inexpensive, noninvasive assessment of the coronary arteries.1 Unlike other imaging modalities, CTCA provides detailed assessment of coronary atherosclerotic plaque with the ability to move beyond traditional assessments of luminal stenosis and to directly phenotype and quantify the burden of coronary atherosclerosis. This includes assessment of the total coronary plaque burden and also the individual component plaque types. In particular, calcific and noncalcific plaque burdens, the latter comprising fibrous plaque, fibrofatty plaque, and low-attenuation plaque as a marker of necrotic core. This advance has been possible only recently with the development of advanced image analysis software. Whereas the published data remain in their early stages, it appears that these assessments, in particular the noncalcific low-attenuation plaque burden, provide powerful prognostic information outperforming other traditional methods.2 To put it simply, the more (noncalcific, necrotic) plaques present, the more likely 1 will rupture and cause a myocardial infarction.
Assessments of atherosclerotic disease activity also hold promise in refining risk prediction. Molecular imaging with positron emission tomography can, in principle, assess the activity of any disease process noninvasively in the body, with novel tracers now allowing investigation of inflammation, calcification, and thrombus formation.3,4 However, CTCA also allows the assessment of coronary inflammation, not by looking at the plaques themselves but by looking at the adjacent pericoronary fat.
Adipose tissue has an important physiological role in maintaining vascular and endothelial function via the action of cytokines named adipokines. The relationship between the vascular endothelium and perivascular adipose tissue (PVAT) is bidirectional. Influenced by local factors such as the vascular redox state and inflammation, or more systemic factors such as obesity and insulin resistance, adipokines can be secreted by the fat to either have pro-oxidative or antioxidative and inflammatory effects on the vasculature.5 Conversely, damage to the vascular endothelium and the resulting pro-inflammatory cytokine release can result in alterations in PVAT composition. The phenotypic change of PVAT can be visualized on CT. In their healthy physiological state, perivascular adipocytes mature into large lipid-laden cells, leading to low attenuation or more negative Hounsfield units measurement (approximately −190 HU) on CT. The presence of inflammation blocks this maturation process, resulting in smaller, aqueous adipocytes, with a less negative Hounsfield unit measurement (up to −30 HU).6
The conceptual advantage of this approach is that this information can be collected on routinely acquired CT scans and therefore does not require the administration of radioactive tracers or access to state-of-the-art molecular positron emission tomography scanners. This facilitates the retrospective assessment of existing data sets, which has greatly accelerated our understanding of the field. Indeed, we now have outcome data performed across multiple studies and different patient populations, which suggests that this approach may refine cardiovascular risk prediction.7,8
It is on this background that Lee et al9 have investigated the association between the change in vessel inflammation, as quantified by PVAT density, and the progression of coronary atherosclerosis. In a retrospective analysis of CTCA scans performed on patients enrolled into the PARADIGM (Progression of AtheRosclerotic PlAque Determined by Computed TomoGraphic Angiography Imaging) registry, the authors analyzed a total of 1,474 lesions from 474 patients. This was a low-risk population; most of the lesions were nonobstructive (mean diameter stenosis <20%), and 14% exhibited high-risk plaque features. The interscan interval between coronary CTAs was 3.3 ± 1.2 years.
PVAT, the total coronary plaque volume, and the volumes of calcific and noncalcific plaque all increased numerically during the study follow-up period, although changes in PVAT were small (−74.1 ± 11.5 HU at baseline vs −73.1 ± 11.7 HU at follow-up). PVAT density increased in 788 lesions and decreased in 688 lesions over time. At follow-up, the lesions with higher (less negative) PVAT density possessed a greater total plaque volume and a larger volume of noncalcified plaque than lesions with lower PVAT density.
The headline result was that on multivariable analysis, the observed increase in PVAT density demonstrated a positive and independent association with the progression of the total coronary plaque volume. Given that inflammation is 1 of the primary mechanisms responsible for the development and progression of coronary atherosclerosis,10 this observation makes intuitive sense.
Associations between changes in PVAT and changes in the different subtypes of coronary plaque were less straightforward. No association was observed between changes in PVAT and changes in calcific plaque, which was perhaps expected, given the stable nature of this plaque phenotype. By contrast, there was an observed trend (P = 0.053) between change in PVAT with changes in the noncalcific plaques that were more closely aligned with inflammation and cardiovascular events. So far so good. However, no significant correlation was observed between changes in PVAT and changes in low-attenuation plaque. This is less intuitive, given that low-attenuation plaque quantifies necrotic core and therefore conceptually represents the most inflamed plaque phenotype. Similarly, there was no relationship with changes in the other inflammatory plaque phenotype: the fibrofatty plaque. Instead, the closest association was observed between change in PVAT density and progression of fibrous plaque. This is somewhat unexpected, given that fibrous plaque is generally considered a less inflamed plaque phenotype. Further studies are now required to investigate these observations and the relationship between PVAT and the different plaque phenotypes in more detail.
It is important to note some important limitations in this study. There was no histologic validation to confirm whether changes in PVAT density truly reflect changes in plaque inflammation. There was inconsistent statin use among the participants, which may have confounded some of the observations. There was also a selection bias against the recruitment of higher-risk patients with the most rapid disease progression. Finally, given the low-risk, relatively small population, this study was not powered to assess whether the observed changes in PVAT or plaque volume are associated with adverse cardiovascular events. The clinical significance of these longitudinal assessments therefore remains unclear.
In summary, the authors are to be congratulated on this study, which has evaluated the latest advances in CT image analysis within the context of a large international multicenter registry. Consistently with other studies, they have demonstrated the feasibility of tracking changes in coronary plaque phenotype and burden with time and have also demonstrated associations with serial PVAT assessments. Further studies are now required to better understand the link between coronary inflammation and progression of the different plaque types.
Funding Support and Author Disclosures
Dr Dweck has received support from the British Heart Foundation (FS/SCRF/21/32010) and has received the Sir Jules Thorn Award for Biomedical Research 2015 (15/JTA). Dr Loganath has reported that she has no relationships relevant to this paper to disclose.
1. "Computed tomography and cardiac magnetic resonance in ischemic heart disease". J Am Coll Cardiol 2016;68:20: 2201-2216.
2. "Low-attenuation noncalcified plaque on coronary computed tomography angiography predicts myocardial infarction". Circulation 2020;141:18: 1452-1462.
3. "Coronary 18F-sodium fluoride uptake predicts outcomes in patients with coronary artery disease". J Am Coll Cardiol 2020;75:24: 3061-3074.
4. "Detection of atherosclerotic inflammation by Ga-DOTATATE PET compared to FDG PET imaging". J Am Coll Cardiol 2017;69:14: 1774-1791.
5. "Perivascular adipose tissue and coronary atherosclerosis". Heart 2018;104:20: 1654-1662.
6. "Detecting coronary inflammation with perivascular fat attenuation imaging". J Am Coll Cardiol Img 2019;12:10: 2011-2014.
7. "Pericoronary adipose tissue attenuation, low-attenuation plaque burden, and 5-year risk of myocardial infarction". J Am Coll Cardiol Img 2022;15:6: 1078-1088.
8. "Non-invasive detection of coronary inflammation using computed tomography and prediction of residual cardiovascular risk (the CRISP CT study): a post-hoc analysis of prospective outcome data". Lancet 2018;392:10151: 929-939.
9. "Association between changes in perivascular adipose tissue density and plaque progression". J Am Coll Cardiol Img 2022;15:10: XXX-XXX.
10. "Atherosclerosis: an inflammatory disease". N Engl J Med 1999;340:2: 115-126.
The authors attest they are in compliance with human studies committees and animal welfare regulations of the authors’ institutions and Food and Drug Administration guidelines, including patient consent where appropriate. For more information, visit the Author Center.