Perivascular fat imaging by computed tomography (CT): a virtual guide

Figures

• 

  FIGURE 1. Presentation of the four different adipose tissue depots within the thorax from an axial cross‐sectional slice taken from a coronary CT angiogram image

• 

  FIGURE 2. Main roles of key adipocytokines secreted by perivascular adipose tissue (PVAT). IR, insulin resistance; RAAS, renin–angiotensin–aldosterone system

• 

  FIGURE 3. WNT5A is released from perivascular adipose tissue (PVAT) and binds to frizzled receptors on the arterial wall, activating the noncanonical wingless‐related integration site (WNT) signalling pathway. This leads to a USP17/RAC1‐mediated increase in NADPH oxidase activity and superoxide generation and induces endothelial dysfunction via endothelial NOS (eNOS) uncoupling

• 

  FIGURE 4. Illustration of the major components of the bidirectional interplay between the vascular wall and fat in the perivascular space. EC, endothelial cell; eNOS, endothelial NOS; PVAT, perivascular adipose tissue; VSMC, vascular smooth muscle cell

• 

  FIGURE 5. Enhanced production of superoxide leads to increased local production of lipid peroxidation products such as 4‐hydroxynonenal (4‐HNE), which travel into perivascular adipose tissue (PVAT) and up‐regulate adiponectin's gene expression via a PPARγ‐mediated mechanism. Adiponectin, in turn, is secreted and acts on vascular smooth muscle cells and endothelial cells by inhibiting the activity of NADPH oxidases as well as through Akt‐mediated stimulation of endothelial NOS (eNOS) activity and coupling, resulting in increased production of NO and reduced oxidative stress. ADIPOQ, adiponectin; BH2, dihydrobiopterin; BH4, tetrahydrobiopterin; O2 −, superoxide

• 

  FIGURE 6. Morphological appearance of perivascular adipose tissue surrounding healthy and inflamed coronary arteries, depicting the gradient of adipocyte size induced by exogenous vascular inflammation

• 

  FIGURE 7. Illustrative visualisation of the perivascular fat attenuation index (FAI) gradient and radial distance from vascular wall in patients with coronary artery disease (CAD). PVAT, perivascular adipose tissue; RCA, right coronary artery. Parts reproduced with permission from Antonopoulos et al. (2017)

• 

  FIGURE 8. (a) Representative images of delineated perivascular fat with fat attenuation index (FAI) coloured mapping around (i) a culprit lesion (green arrows) demonstrating abnormal FAI in a patient with acute myocardial infarction (MI) (upper left image), (ii) a nonculprit lesion (green arrows) from the same patient also showing abnormal FAI (upper right image), (iii) a stable atherosclerotic lesion without a stent (single green arrow, lower left image) and (iv) a stent (green arrows) implanted at least 3 months before imaging (lower right image); black colour indicates nonadipose tissue with attenuation values outside the −30 to −190‐HU range. (b) Perivascular FAI changes around ruptured (culprit) atherosclerotic lesions (n = 10) of patients with acute MI, nonculprit lesions of the same patients (n = 7) or lesions of stable coronary artery disease (CAD) patients (n = 13). (c, d) Δ[FAI] comparison between stable and unstable plaques and receiver operating characteristic (ROC) curve analysis for its diagnostic accuracy in the detection of unstable plaques (culprit lesions). HU, Hounsfield units; Δ[FAI], FAI (around lesion) − FAI (proximal segment). Reproduced with permission from…

• 

  FIGURE 9. Changes in perivascular fat attenuation index (FAI) in patients with psoriasis not receiving any systemic or biological therapy during 1 year or receiving biological therapy (P = 0.99) during 1 year with anti‐TNF‐α therapy (P < 0.001) or anti‐IL‐12/23 and anti‐IL‐17 therapies (P < 0.001). HU, Hounsfield units. Reproduced with permission from Elnabawi, Oikonomou, et al. (2019). American Medical Association. All rights reserved. Colour images are available online

• 

  FIGURE 10. Kaplan–Meier curves of cardiac mortality with high versus low perivascular fat attenuation index (FAI) for the derivation and validation cohorts. Hazard ratios (HRs) are adjusted for risk factors, technical factors, the extent of coronary artery disease and number of high‐risk plaque features. Comparison of time‐dependent ROC curves (at 6 years) and respective area under the curve (AUC) of two nested models for discrimination of cardiac mortality in the derivation and validation cohorts. Model 1 represents the current state of the art in risk assessment and consists of age, sex, risk factors (hypertension, hypercholesterolaemia, diabetes mellitus, smoker status and epicardial adipose tissue volume), modified Duke coronary artery disease index and number of high‐risk plaque features on coronary CTA. Model 2 incorporates perivascular FAI values into Model 1. HRP, high‐risk plaque; HU, Hounsfield units. Reproduced with permission from Oikonomou et al. (2018)

• 

  FIGURE 11. Subgroup analysis based on clinical recommendation for treatment with statin and/or aspirin after coronary CT angiography (CCTA) about the predictive value of perivascular fat attenuation index (FAI) for all‐cause (a) and cardiac mortality (b). Cox regression models were adjusted for age, sex and epicardial adipose tissue volume. FAI was strongly associated with cardiac mortality events in patients that did not receive recommendations for treatment with statins or aspirin, whereas the association in the group that did receive such recommendations was non‐significant. This highlights a potential role for FAI in secondary prevention measures and personalised selection of patients that can qualify for medical treatment after screening in current clinical practice. CI, confidence interval. Reproduced with permission from Oikonomou et al. (2018)

• 

  FIGURE 12. Vision for implementing fat attenuation index (FAI) and high‐risk plaque (HRP) features in clinical practice. anti‐PCSK9, pro‐protein convertase subtilisin/kexin type 9 inhibitors; CCTA, coronary CT angiography; f/u, follow‐up. *% referred to data from CRISP‐CT study. Reproduced with permission from Antoniades, Kotanidis, and Berman (2019)