Abstract
Objectives:
In stable patients with high-risk coronary plaques on coronary CT angiography (CTA), we aimed to assess the association between increased lesion pericoronary adipose tissue (PCAT) density and coronary 18F- sodium fluoride (18F-NaF) uptake on positron emission tomography (PET).
Background:
Coronary 18F-NaF uptake reflects coronary microcalcification. Increased PCAT density is associated with vascular inflammation. To date the relationship between increased PCAT density and ¹⁸F-NaF uptake in stable patients with high risk plaques on coronary CTA has not been characterized.
Methods:
Patients undergoing coronary CTA were screened for HRP defined by 3 concurrent plaque features: positive remodeling; low attenuation plaque (LAP, <30 HU), spotty calcification; obstructive coronary stenosis ≥50%; plaque volume >100mm3. Patients with HRPs were recruited for 18F-NaF PET/CT. In lesions with stenosis ≥25%, quantitative plaque analysis, mean PCAT density, maximal coronary motion-corrected 18F-NaF standard uptake values (SUVmax) and target to background ratios (TBR) were measured.
Results:
Forty-one patients (age 65±6 years, 68% male) were recruited. Fifty-one lesions in 23 patients (56%) showed increased coronary 18F-NaF activity. Lesions with 18F-NaF uptake had higher surrounding PCAT density than those without (−73 [interquartile range −79 to −68] vs. −86 [−94 to −80] HU, p<0.001).
18F-NaF TBR and SUVmax correlated with PCAT density (r=0.63 and r=0.68, all p<0.001). On adjusted multiple regression analysis, increased lesion PCAT density and LAP volume were associated with 18F-NaF TBR (β=0.25 95% CI 0.17–0.34, p<0.001 for PCAT and β=0.07 95% CI 0.03–0.11, p=0.002 for LAP).
Conclusions:
In patients with high-risk plaque features on coronary CTA, increased density of PCAT is associated with focal 18F-NaF PET uptake. Simultaneous assessment of these imaging biomarkers by 18F-NaF PET and CTA may refine CAD assessment in stable patients with high-risk plaque features.
Keywords: PET/CT, Coronary computed tomography angiography, Coronary imaging, High risk plaque, Pericoronary adipose tissue density, 18F-sodium fluoride
Introduction
Pathological studies have identified adverse plaque features of ruptured coronary lesions. These include a large lipid core, spotty calcification, positive remodeling and inflammatory cell infiltration (1,2). Since most of these plaque features can be readily detected on coronary computed tomography angiography (CTA), multiple studies have evaluated the impact of such lesion characteristics on patient outcome. Initial studies showed promise in this regard, and as a result, study of high-risk plaque (HRP) on CTA has gained great interest (3). In a recent study, however, while adverse plaque features have been shown to be predictors of acute coronary syndromes, the cumulative number of patients implicated in acute myocardial infarction with and without such morphological characteristics was similar (4). Indeed, the absolute differences in event rates in those with and without adverse plaque features are low and the positive predictive value of adverse plaque features for myocardial infarction is very low (5,6).
This low positive predictive value of adverse plaque features mandates further research aimed at distinguishing subjects at high risk of adverse events. Recently, 18F-sodium fluoride PET/CT (18F-NaF) and increased pericoronary adipose tissue (PCAT) CT attenuation have emerged as potential markers in this regard. ¹⁸F-NaF uptake has been reported in ruptured plaques in patients with acute myocardial infarction (7). Similarly, increased density of PCAT has recently been shown to provide insight into vascular inflammation, which drives the development and progression of coronary atherosclerosis (8–11). Furthermore, increased PCAT density co-localizes with culprit plaques in patients with acute coronary syndromes (12). To date the relationship between increased PCAT density and ¹⁸F-NaF uptake in stable patients with HRPs on coronary CTA has not been characterized.
The aim of this study, therefore, was to investigate the relationship between novel imaging approaches (18F-NaF uptake and CT attenuation of the pericoronary adipose tissue density) in stable patients with HRP.
Methods
Patients
At two high-volume cardiac imaging centers (Cedars-Sinai Medical Center & Severance Hospital, Yonsei University College of Medicine), stable patients with borderline coronary stenosis in the main epicardial vessels on CTA were screened for HRP by visual and quantitative plaque analysis, as described below.
Inclusion Criteria
Patient inclusion required a coronary HRP defined by having at least three of the following adverse plaque features by quantitative plaque analysis of CTA performed for clinical purposes in stable subjects: 1) positive remodeling (remodeling index >1.1); 2) low attenuation plaque (LAP) (<30 HU); 3) obstructive coronary stenosis defined by a vessel stenosis ≥50% or contrast density difference (the difference between the lumen opacity proximal and distal to the plaque) ≥25% (13); 4) plaque volume >100mm3 or plaque burden (lesion volume normalized to the vessel volume) ≥40%; 5) spotty calcification on visual CTA assessment.
Those who displayed HRP on CTA were invited to participate in a research 18F-NaF PET-CT scan. All recruited subjects underwent a comprehensive baseline clinical assessment, including evaluation of their lipid and cardiovascular risk factor profiles and 18F-NaF PET/CT imaging. The study was approved by local institutional review boards and all patients provided written informed consent.
Exclusion Criteria
We excluded patients younger than 18 years, with elevated creatinine>1.5 mg/dL, pregnant or breastfeeding females, subjects with significant arrhythmia including multiple premature ventricular or atrial contractions, patients with an ejection fraction <35% or class III congestive heart failure. Finally, individuals who had a known iodine contrast allergy, were unable to take beta blockers or were unable to provide informed consent were also excluded.
Coronary Computed Tomography Angiography
Imaging was performed using a dual source CT scanner (Definition; Siemens Healthcare, Forchheim, Germany) at Cedars-Sinai Medical Center and 320-slice CT scanner (Aquilion One, Toshiba Medical Systems, Tokyo, Japan) at Severance Hospital using standard protocols (12). Prior to the acquisition of the CTA all patients underwent a non-contrast CT scan for coronary calcium assessment. All patients received betablockers (to achieve a target heart rate <60 beats/min), sublingual nitrates, and power-injected iodinated contrast (350mgI/mL, 65–130 mL).
Plaque analysis
Visual evaluation of coronary stenosis, positive remodeling, spotty calcification and segment involvement score (SIS) was conducted by at least two experienced readers in accordance with the Society of Cardiovascular Computed Tomography guidelines (14,15). Multivessel coronary artery disease was defined as at least two major epicardial vessels with >50% luminal stenosis. The presence of LAP was determined by the reader at the time of initial reading using software from the scanner manufacturer (Vitrea, Vital Images, Minnetonka, Minnesota or SyngoVia, Siemens, Erlangen, Germany) to define regions within the plaque of interest of <30 HU.
Quantitative plaque analysis of all coronary segments with a lumen diameter ≥2 mm was performed using semi-automated software (AutoPlaque version 2.0, Cedars-Sinai Medical Center, Los Angeles, USA; 13). Coronary CTA images were examined in multiplanar format, and proximal and distal limits of the plaque lesions were manually marked by an experienced reader. Subsequent plaque quantification was fully automated using adaptive scan-specific thresholds. Total (TPV), calcified, non-calcified (NCP) as well as LAP volumes were calculated in mm3. The plaque burden was calculated according to the following equation (plaque volume x 100%/vessel volume). Positive remodeling was defined as a remodeling index exceeding 1.1, the latter expressed as vessel cross sectional area at the lesion divided by the reference cross-sectional area. The contrast density difference was the maximal difference in contrast density in the plaque and the reference proximal vessel cross section. The Agatston coronary calcium score was measured using semiautomatic commercial software (NetraMD, ScImage, Los Altos, California).
Pericoronary adipose tissue analysis
To measure PCAT density, three-dimensional layers within radial distance from the outer coronary wall equal in thickness to the average diameter of the vessel were constructed automatically from the CTA (Figures 2 and 4) using semi-automated software (AutoPlaque version 2.0, Cedars-Sinai Medical Center, Los Angeles, USA). Within the predefined volume of interest, voxels with tissue attenuation ranging from −190 up to −30 HU were considered as adipose tissue. Neither myocardial tissue adjacent to the vessel wall nor the coronary branches originating from the vessel of interest were included in the automated analysis. Pericoronary adipose tissue density was defined as the mean attenuation within such contamination-free volumes of interest. These measurements were performed in all coronary lesions as well as in a reference region of the PCAT surrounding the proximal and mid right coronary artery, as described recently (8).
Figure 2. Case illustrations of coronary CTA and assessment of 18F-NaF uptake and PCAT attenuation in high-risk plaques.
Patient 1: 53-year-old male with a RCA plaque with positive remodeling (green arrow) (A), focal 18F-NaF uptake with increased target to background ratio (TBR) of 1.73 (B), and increased PCAT attenuation (mean PCAT density [−76.7 HU]) (C). Patient 2: 66-year-old male with a LAD lesion with low attenuation plaque (green arrow) (E), focal 18F-NaF uptake with increased TBR of 1.87 (F), and increased PCAT attenuation (mean PCAT density of [−74.8 HU] (F). Patient 3: 54-year-old male with a LAD lesion with low attenuation plaque (green arrow) (G), focal 18F-NaF uptake with increased TBR of 2.28 (H), and increased PCAT attenuation (mean PCAT density [−73.6 HU]) (I). HU - Hounsfield Units; LAD - left anterior descending; PCAT - pericoronary adipose tissue; RCA - right coronary artery; SUVmax -maximum standard uptake values; TBR - target to background ratio
Figure 4. Pericoronary adipose tissue density evaluation of plaques in the LAD (A-C) and RCA (D-F).
Multiplanar reconstructed (A, D), cross-sectional (B, E) images and histograms presenting pericoronary adipose tissue assessment. Cylindrical volumes of interest are drawn automatically radially from the vessel wall within the distance equal to the artery diameter (red, orange, yellow on MPR [A, D] and cross-sectional [B, E] images). Only pixels with attenuation: −190 to −30 are considered as adipose tissue. Myocardium adjacent to the vessel wall and coronary branches were automatically excluded (panels A, B and D, E). The mean adipose tissue density within the entire volume of interest was used for analysis. LAD - left anterior descending; PCAT – pericoronary adipose tissue RCA - right coronary artery
18F-Sodium Fluoride PET/CT
All patients underwent 18F-NaF PET/CT on a hybrid PET-CT scanner (Discovery 710, GE Healthcare, Milwaukee, WI, USA). Prior to imaging, subjects were administered with a target dose of 250 MBq of 18F-NaF and rested in a quiet environment for 60 min. After the acquisition of a non-contrast attenuation correction scan, PET data was acquired in list mode for 30 min. In the final step, a CTA was performed. The electrocardiography-gated list mode PET/CT dataset was reconstructed using a standard ordered expectation maximization algorithm with time of flight and resolution recovery (SharpIR). Utilizing 10 cardiac gates, we reconstructed the data on a 256×256 matrix with 4 iterations and 5mm Gaussian smoothing (16).
Motion correction
Cardiac motion correction was performed for the PET/CT images. This novel technique compensates for coronary artery motion by aligning all gates to the end-diastolic position, thus allowing utilization of radioactivity throughout the cardiac cycle. This method has been shown to reduce image noise and improve target to background ratios (17). In the first step of motion correction, anatomic coronary artery data was extracted from coronary CTA by applying a vessel tracking algorithm based on Bayesian maximal paths (Autoplaque version 2.0). Second, a diffeomorphic mass-preserving image registration algorithm was used to align the 10 gates of PET data to the end-diastolic gate (FusionQuant Software, Cedars Sinai Medical Center, Los Angeles). After motion correction, the 10 gates were summed together to build a motion-free image containing counts from the entire PET acquisition.
Image analysis
Coronary 18F-NaF image analysis was performed on axial images using FusionQuant software. PET and CTA reconstructions were reoriented, fused and systematically co-registered in all 3 planes. Key points of reference were the sternum, vertebrae, blood pool in the ventricles and the great vessels. For each scan, plaque activity was measured by delimiting 3-dimensional volumes of interest on lesions. On motion corrected images all segments with coronary plaque (at least a >25% stenosis) that were suitable for Autoplaque analysis (had a lumen diameter ≥2 mm) were interrogated. The maximum standard uptake value (SUVmax) was recorded from these regions. Background blood pool activity was measured by delimiting a cylindrical volume of interest (radius = 10 mm, thickness = 5 mm) in the right atrium on the level of the right coronary artery orifice. Target to background ratios (TBRs) were calculated by dividing SUVmax by averaged background blood pool activity. Coronary plaques were considered positive for 18F-NaF if they presented with focal tracer uptake arising from the coronary plaque which followed the course of the vessel in three dimensions over more than one slice and had a TBR>1.25.
Statistical Analysis
Data were tested for normality using Shapiro Wilks test. Continuous data are expressed as mean (standard deviation) or median [interquartile range (IQR)] dependent on the distribution. Skewed continuous variables were log transformed to achieve normal distribution. Data were compared with the two-sample t-test and the relationship between 2 continuous variables was assessed using Pearson’s r. Categorical variables are presented as absolute numbers (percentage) and were compared using a Chi-squared test. Multivariate logistic regression and linear regression modelling were used to assess the change in CT derived markers of coronary artery disease, clinical characteristics and 18F-NaF uptake. In the per lesion analysis no adjustments were made for multiple observations (plaques) within a single patient or vessel territory. A 2-sided p value <0.05 was regarded as significant. Further statistical analysis details including power calculations are presented in the online supplement. Statistical analysis was performed with SPSS software (version 24, SPSS, Inc., Chicago, Illinois).
Results
Patient population
Forty-one patients (age 65±6 years, 68% male) were recruited into the study (Table 1). All individuals underwent PET/CT within 119 days (median 17 [IQR, 14 to 37]) from the initial CTA acquisition. There were no differences between sites in baseline characteristics, co-morbidities, risk factors, medications and presenting complaint (all p>0.1). There were no significant differences noted in these characteristics between the sites (Online Table 1).
Table 1:
Baseline characteristics of patient who comprised the study population.
| Study cohort (n=41) | |
|---|---|
| Baseline Characteristics | |
| Age, years | 65±6 |
| Males, n (%) | 28 (68%) |
| Diabetes, n (%) | 6 (14%) |
| Hyperlipidemia, n (%) | 22 (54%) |
| Hypertension, n (%) | 16 (39%) |
| Tobacco use, n (%) | 7 (17%) |
| Family history of CAD, n (%) | 13 (32%) |
| Serum Biomarkers | |
| Total Cholesterol, mg/dL | 166 [152, 194] |
| High Density Lipoprotein, mg/dL | 44 [39,50] |
| Low Density Lipoprotein, mg/dL | 98 [70, 120] |
| Triglyceride, mg/dL | 118 [87, 138] |
| Creatine, mg/dL | 0.9 [0.7, 1.0] |
| Medications | |
| Aspirin, n (%) | 29 (71%) |
| Statin, n (%) | 19 (46%) |
| ACEI/ARB, n (%) | 9 (22%) |
| Beta Blocker, n (%) | 17 (41%) |
| Leading clinical indication for CTA | |
| Chest pain, n (%) | 22 (54%) |
| Dyspnea, n (%) | 6 (15%) |
| Risk assessment (asymptomatic patient), n (%) | 13 (32%) |
| Coronary Computed Tomography Angiography | |
| Segment involvement score | 6 [5, 9] |
| Multivessel disease, n (%) | 15 (37%) |
| Coronary calcium score | 406 [150, 787] |
| Total plaque volume (mm3) | 1261 [745, 2298] |
| NCP volume (mm3) | 1141 [661, 1931] |
| median PCAT attenuation, HU | −81 [−88, 75] |
| PET/CT | |
| Patient with lesion uptake, n (%) | 23 (56%) |
| Segments with uptake, n (%) | 51 (8%) |
| SUV max | 1.6 [1.2, 2.2] |
| TBR lesions with uptake | 1.6 [1.3, 1.8] |
| TBR coronary lesions | 1.2 [0.9, 1.5] |
ACEI – angiotensin converting enzyme inhibitor; ARB – angiotensin receptor blocker; CAD: coronary artery disease; CCTA - coronary computed tomography angiography; HU – Hounsfield units; NCP – non-calcified plaque; PCAT – Pericoronary adipose tissue; SUVmax - maximum standard uptake values; TBR - target to background ratio;
Coronary Computed Tomography Angiography
Forty-seven lesions that met HRP criteria were identified in the 41 patients. Of the predefined adverse features, plaque volume >100 mm3 or plaque burden ≥40% and visual stenosis ≥50% or a contrast density difference ≥25% were detected in 37 (79%) and 39 (83%) lesions respectively. Positive remodeling was found in 32 (68%) cases and LAP (<30 HU) was observed in 27 (57%) of the HRPs. Spotty calcification was detected in 6 (13%) lesions.
The median pericoronary adipose tissue density surrounding coronary segments with at least 25% stenosis (n=132) was −81 [IQR, −88 to −75] HU and showed no difference compared to the median PCAT density surrounding the proximal RCA −79 [IQR, −86 to −73] HU, p=0.45. The HRPs (n=47) had a median PCAT density of −78 [IQR, −82 to −70] HU compared to −81 [IQR, −85 to −74] HU in the remaining lesions (n=85; p=0.23).
18F-NaF PET/CT
On visual assessment, twenty-three (56%) patients showed 18F-NaF uptake in the coronary vasculature, and 51 coronary segments with increased PET tracer activity were identified (Figures 2 and 3). Twenty-eight (59%) of the 47 HRPs showed 18F-NaF uptake. The median SUVmax and TBR in lesions with uptake were respectively 2.2 [IQR, 1.8 to 3.1] and 1.6 [IQR, 1.3 to 1.8] compared to 1.3 [IQR, 1.0 to 1.6] and 1.0 [IQR, 0.7 to 1.1] in those without increased tracer activity.
Figure 3. Examples of 18F-NaF uptake in six patients with high risk plaque characteristics.
Focal uptake in the proximal LAD (A, B), left main and LAD (C), proximal (D) and distal (E) RCA and proximal LCX. LAD - left anterior descending; LCX - left circumflex; RCA - right coronary artery
There were no differences in lesion location, degree of stenosis, total and non-calcified plaque volumes between lesions with and without 18F-NaF uptake (all: p>0.10).
Differences in these measures were observed in a per patient analysis (Table 2). Compared to those without 18F-NaF uptake, patients presenting with 18F-NaF uptake had higher segment involvement (7 [IQR, 5–9] vs 4 [IQR, 3–7], p=0.006) and coronary calcium scores (750 [IQR, 406–1059] vs 150 [IQR, 21–306] Agatston units, p=0.008). Moreover, compared to subjects without increased 18F-NaF uptake, these patients had higher total (1723 [IQR, 1070–2523] vs 747 [IQR, 458–1389] mm3, p=0.004) and non-calcified (1614 [IQR, 951–2161] vs 661 [IQR, 456–1238] mm3, p=0.004) plaque volumes.
Table 2.
Comparison of patients with and without 18F-NaF uptake.
| 18F-NaF uptake (n=23) | No 18F-NaF uptake (n=18) | P value | |
|---|---|---|---|
| Baseline Characteristics | |||
| Age, years | 68 [62, 71] | 65 [57, 69] | 0.91 |
| Males, n (%) | 15 (65%) | 13 (72%) | 0.79 |
| Diabetes, n (%) | 3 (13%) | 3 (17%) | 0.92 |
| Hyperlipidemia, n (%) | 12 (52%) | 10 (56%) | 0.79 |
| Hypertension, n (%) | 8 (35%) | 8 (44%) | 0.38 |
| Tobacco use, n (%) | 4 (17%) | 3 (17%) | 0.74 |
| Family history of CAD, n (%) | 6 (26%) | 7 (39%) | 0.79 |
| ASCVD score, % | 13.6 [11.4, 26.9] | 13.8 [8.7, 27.4] | 0.95 |
| Serum Biomarkers | |||
| Total Cholesterol | 171 [138, 202] | 154 [129, 181] | 0.71 |
| High Density Lipoprotein | 43 [34, 53] | 46 [39, 54] | 0.16 |
| Low Density Lipoprotein | 93 [64, 124] | 89 [66, 101] | 0.97 |
| Triglyceride | 103 [82, 142] | 87 [66, 123] | 0.27 |
| Creatine | 0.8 [0.7, 0.9] | 0.8 [0.7, 0.9] | 0.13 |
| Medications | |||
| Aspirin, n (%) | 18 (78%) | 11 (61%) | 0.86 |
| Clopidogrel, n (%) | 13 (57%) | 7 (39%) | 0.79 |
| Statin, n (%) | 10 (43%) | 9 (50%) | 0.93 |
| ACEI/ARB, n (%) | 7 (30%) | 2 (11%) | 0.25 |
| Beta Blocker, n (%) | 10 (43%) | 7 (39%) | 0.86 |
| Leading clinical indication for CTA | |||
| Chest pain, n (%) | 12 (53%) | 10 (56%) | 0.83 |
| Dyspnea, n (%) | 4 (17%) | 2 (11%) | 0.66 |
| Risk assessment (asymptomatic patient), n (%) | 7 (30%) | 6 (33%) | 0.50 |
| Coronary Computed Tomography Angiography | |||
| Segment involvement score | 7 [5, 9] | 4 [3, 7] | 0.006 |
| Multivessel disease, n (%) | 10 (43%) | 5 (28%) | 0.24 |
| Coronary calcium score | 750 [406, 1059] | 150 [0, 306] | 0.008 |
| Total plaque volume (mm3) | 1723 [1070, 2523] | 747 [458, 1389] | 0.004 |
| NCP volume (mm3) | 1614 [951, 2161] | 661 [456, 1238] | 0.004 |
| Median PCAT density, HU | −80 [−89, −71] | −83 [−93, −75] | 0.16 |
| High risk plaque features (patients with) | |||
| Low attenuation plaque, n (%) | 15 (65%) | 4 (22%) | 0.025 |
| Positive remodeling, n (%) | 17 (74%) | 13 (72%) | 0.74 |
| Spotty Calcification, n (%) | 6 (26%) | 3 (17%) | 0.71 |
| Stenosis ≥50% or contrast density difference ≥25%, n (%) | 20 (87%) | 17 (94%) | 0.88 |
| Plaque volume >100mm3 or plaque burden ≥40%, n (%) | 20 (87%) | 16 (89%) | 0.91 |
| PET/CT | |||
| SUVmax | 2.3 [1.8, 3.1] | 1.3 [1.0, 1.6] | 0.001 |
| TBRmax | 1.6 [1.3, 1.9] | 1.0 [0.7, 1.1] | 0.003 |
ACEI – angiotensin converting enzyme inhibitor; ARB – angiotensin receptor blocker; ASCVD: Atherosclerotic Cardiovascular Disease risk score; CAD: coronary artery disease; CTA - coronary computed tomography angiography; HU – Hounsfield units; NCP – non-calcified plaque; PCAT – Pericoronary adipose tissue; SUVmax - maximum standard uptake values; TBRmax - maximum target to background ratio;
18F-NaF uptake, PCAT density and LAP
Lesions with increased activity of 18F-NaF had higher surrounding PCAT density than those without (−73 [IQR, −79 to −68] vs. −86 [IQR, −94 to −80] HU, p<0.001; Figure 1A, Table 3). Importantly, there was an association between PET tracer uptake and the density of PCAT surrounding corresponding lesions [r=0.63, p<0.001 for SUVmax and r=0.68, p<0.001 for TBR measurements (Figure 1B&C)]. There were weak correlations between lesion LAP volumes and corresponding PCAT and TBR values (r=0.43, p<0.001 and r=0.45, p<0.001 respectively). When PCAT analyses was confined to the proximal and mid RCA, there was no difference in PCAT density between patients with and without increased 18F-NaF activity (−81 [IQR, −84 to −74] vs −85 [IQR, −90 to −79] HU, p=0.21).
Figure 1. Pericoronary adipose tissue density and 18F-NaF uptake.
Lesions with PET tracer uptake had higher PCAT density than those without (−73.2 [−78.7, −68.5] vs −86.3 [−93.8, −79.8] HU, p<0.001) (A). There was moderate correlation between lesion PCAT density and SUVmax (r=0.63, p<0.001) (B) and a stronger association between PCAT density and TBR (r=0.68, p<0.001) (C). HU - Hounsfield Units; PCAT - pericoronary adipose tissue; SUVmax -maximum standard uptake values; TBR - target to background ratio
Table 3.
Comparison of lesions with and without 18F-NaF uptake.
| 18F-NaF uptake (n=51) | No 18F-NaF uptake (n=81) | P value | |
|---|---|---|---|
| Lesion localization | |||
| LM, n (%) | 5 (20) | 7 (15) | 0.80 |
| LAD, n (%) | 9 (36) | 19 (40) | 0.42 |
| LCX, n (%) | 5 (20) | 9 (19) | 0.66 |
| RCA, n (%) | 6 (24) | 12 (26) | 0.58 |
| Coronary Computed Tomography Angiography | |||
| Lesion stenosis, % | 51 [35, 75] | 48 [36, 65] | 0.71 |
| Total plaque volume, mm3 | 219 [136, 391] | 197 [122, 307] | 0.27 |
| NCP volume, mm3 | 199 [132, 339] | 172 [119, 272] | 0.18 |
| Low attenuation plaque volume, mm3 | 48 [37, 66] | 19 [12, 26] | <0.001 |
| Lesion PCAT, HU | −73 [−79, −68] | −86 [−94, −80] | <0.001 |
| PET/CT | |||
| SUVmax | 2.2 [1.8, 3.1] | 1.3 [1.0, 1.6] | <0.001 |
| TBR | 1.6 [1.3, 1.8] | 1.0 [0.7, 1.1] | <0.001 |
LM- left main; LAD - left anterior descending; LCX - left circumflex; RCA - right coronary artery; NCP - non-calcified plaque; HU - Hounsfield Units; PCAT - pericoronary adipose tissue; SUVmax -maximum standard uptake values; TBR - target to background ratio;
Multiple Regression Analysis.
In an adjusted multivariable linear regression analyses, lesion PCAT density and LAP volume were associated with 18F-NaF TBR: β=0.25 per 11 HU PCAT density increase (95% CI 0.17 to 0.34, p<0.001) and β=0.07 per 14mm3 LAP volume increase (95% CI 0.03 to 0.11, p=0.002) after adjustments for lesion quantitative percent stenosis, total and non-calcified plaque volumes (Table 4). Concordant associations were found on multivariable logistic regression analysis (Online Table 2).
Table 4.
Univariable and multivariable linear regression analysis of plaque characteristics for lesion TBR increase. Beta coefficients are expressed per 1 SD increment.
| Plaque characteristic | Univariate | Multivariable | ||
|---|---|---|---|---|
| B (95% CI) | P value | B (95% CI) | P value | |
| Total volume, per 38mm3 | 0.0037 (−0.004–0.0077) | 0.48 | −0.01 (−0.14–0.13) | 0.81 |
| NCP volume, per 22mm3 | 0.034 (−0.002–0.08) | 0.28 | 0.02 (−0.16–0.19) | 0.64 |
| Lesion stenosis, per 21% | −0.04 (−0.112–0.016) | 0.16 | −0.04 (−0.10–0.021) | 0.23 |
| Low attenuation plaque, per 14mm3 | 0.12 (0.07–0.18) | <0.001 | 0.07 (0.03–0.11) | 0.002 |
| Lesion PCAT density, per 11HU | 0.31 (0.20–0.43) | <0.001 | 0.25 (0.17–0.34) | <0.001 |
CI - confidence interval; HU – Hounsfield units; NCP - non-calcified plaque; PCAT - pericoronary adipose tissue; TBR - target to background ratio;
Similarly, after adjustments for quantitative percent lesion stenosis, as well as total and non-calcified plaque volumes in a multivariable linear regression analyses, lesion 18F-NaF TBR and LAP volume were associated with lesion PCAT: β=6.3 per 0.47 TBR increase (95% CI 4.6 to 7.8, p<0.001) and β=0.84 per 14 mm3 LAP volume increase (95% CI 0.02 to 1.66, p=0.029; Table 5).
Table 5.
Univariable and multivariable linear regression analysis of plaque characteristics for lesion PCAT density increase. Beta coefficients are expressed per 1 SD increment.
| Plaque characteristic | Univariate | Multivariable | ||
|---|---|---|---|---|
| B (95% CI) | P value | B (95% CI) | P value | |
| Total volume, per 38mm3 | 0.32 (−0.04–0.70) | 0.18 | −0.38 (−1.9–1.1) | 0.58 |
| NCP volume, per 20mm3 | 0.45 (−0.04–0.87) | 0.12 | 0.4 (−0.81–1.6) | 0.31 |
| Lesion stenosis, 21% | −0.17 (−3.36–0.18) | 0.09 | −1.02 (−2.42–0.38) | 0.14 |
| Low attenuation plaque, per 14mm3 | 2.65 (1.69–3.75) | <0.001 | 0.84 (0.02–1.66) | 0.029 |
| Lesion TBR, per 0.47 | 7.3 (5.9–8.7) | <0.001 | 6.3 (4.6–7.8) | <0.001 |
CI - confidence interval; NCP - non-calcified plaque; PCAT - pericoronary adipose tissue; TBR - target to background ratio;
Discussion
This is the first study to evaluate the relationship between PCAT density on CTA and 18F-NaF uptake by PET in patients with HRP on CTA. Over 50% of patients were found to have increased coronary 18F-NaF activity. Moreover 18F-NaF uptake, whether measured using SUVmax or TBR was associated with lesion PCAT density. While the majority of CT-derived quantitative lesion indices did not distinguish plaques with tracer uptake, lesion PCAT density and LAP volume were independent predictors of increased 18F-NaF coronary activity. Quantitative measures of plaque composition were superior to visual CTA assessment in distinguishing 18F-NaF coronary uptake at a plaque and patient level.
In this study we focused on patients with high risk plaque features because we believe that if the novel imaging approaches for plaque evaluation (PCAT and 18F-NaF) are to be used in clinical practice, a population of patients with adverse plaque features would potentially benefit the most from these assessments. For recruitment patients had to fulfil 3 out of 5 criteria. Four of these have been linked to adverse outcome (6,13) and the inclusion of the stenosis criterium ensures clinical relevance as lesions with an over 50% stenosis can be considered for coronary intervention. Our idea is based on the concept that a strategy using stenosis guided or ischemia guided referral to catheterization and possible revascularization of patients without refractory symptoms is flawed by not considering disease activity.
PCAT assessment has gained considerable interest as a potential biomarker derived from CTA (8–11). In a landmark manuscript, Antonopoulos et al demonstrated through multiple lines of evidence a relationship between coronary perivascular fat attenuation and inflammation in and around the proximal RCA (8). The authors showed that the fat attenuation gradient around human coronary arteries identified early subclinical coronary artery disease in vivo, as well as vulnerable atherosclerotic plaques during acute coronary syndromes. More recently in a retrospective analysis of two large prospectively recruited CTA cohorts it was shown that high PCAT values were predictive of all-cause and cardiac mortality (11). Furthermore, elevated PCAT attenuation was associated with an increased risk of acute myocardial infarction (11). Rather than being limited to the proximal segments of coronary arteries which can be remote from HRP identified on CTA, the method we applied enables assessment of PCAT density adjacent to all coronary lesions. Evaluating plaques in all coronary segments, we found that those with 18F-NaF uptake had higher PCAT density than those without increased tracer activity. This relationship was not observed when only the proximal RCA was assessed for PCAT. These findings suggest that a lesion specific assessment of PCAT might provide greater insight into atherosclerosis biology than RCA measurements alone.
In line with our observations, in a recent study Goeller et al. showed that PCAT density was increased around culprit lesions compared with non-culprit lesions in patients with acute coronary syndromes and stenosed lesions in stable CAD controls (12). Additionally, the authors reported that LAP is independently associated with culprit lesions. Both these findings highlight the potential importance of a lesion specific evaluation of PCAT density and plaque characteristics. The associations between LAP, PCAT and culprit lesions in the prior study (12) and the relationship between 18F-NaF uptake with LAP and PCAT provides potential intriguing mechanistic insights regarding the biology of atherosclerosis.
Increased coronary artery uptake of 18F-NaF occurs in lesions associated with acute coronary syndromes and is considered to reflect not only the presence of microcalcification but also the rate of microcalcification occurring in response to coronary inflammation (7). ln this study, we provide evidence that increased PCAT density, a novel imaging measurement potentially associated with active vascular inflammation and 18F-NaF uptake, potentially reflecting the rate of deposition of calcium in response to inflammation, are closely associated. On the molecular level, the association between osteogenesis and inflammation in atherosclerosis has been widely reported, and our study supports these findings at a non-invasive imaging level (18–20). This association has also been explored in the histological assessment of carotid endarterectomy specimens and preclinical models (8,21,22). The authors found that plaques with increased 18F-NaF uptake had not only increased microcalcification, but also more pronounced macrophage infiltration and apoptosis. The finding of the correlation between PCAT density and 18F-NaF activity supports the link between vascular inflammation and microcalcification at a non-invasive imaging level and suggests that both these imaging approaches provide information regarding plaque activity.
Evaluation of coronary artery plaque by coronary artery calcification or by visual and quantitative assessments of coronary CTA has been shown to be strongly predictive of patient risk and to better guide patient management decisions than standard clinical data (23–25). Anatomic assessment, however, does not provide information regarding the activity of CAD. In this study, we sought to assess novel imaging approaches which may reflect atherosclerosis activity. The finding that not all patients with plaques with adverse features on visual or quantitative assessments on CTA present with 18F-NaF uptake or altered PCAT density suggests that potentially risk assessment can be refined by evaluating atherosclerotic plaque biology by means of measuring microcalcification (18F-NaF uptake) and/or a marker associated with vascular inflammation (increased PCAT density). Potentially, these assessments could help distinguish patients with highly active disease, thereby addressing a need for a more individualized patient management strategy.
Study Limitations
There are several limitations of our work. Due to the modest number of patients that comprised the study cohort, assessment of the predictive value of our findings on a per patient level by means of linear and logistic regression modelling was not feasible. This is particularly important as individuals with 18F-NaF coronary uptake had a higher burden of atherosclerosis (as shown by differences in the segment involvement, coronary calcium scores and total, non-calcified plaque volumes). The proposed risk features should be assessed against the global atherosclerosis burden. Similarly, we have not made adjustments for potentially correlated multiple observations within a single patient. The napkin-ring sign has been described as a high-risk plaque feature was not included in our selection of patients for inclusion in the study as we considered it more subjective than other reported high-risk features. Upon review of the patients included in the study there were no plaques with napkin ring sign plaques. With the definition used, at least one of the three conventional high-risk features—positive remodeling, lipid (<30 HU), and spotty calcification—was required to satisfy inclusion criteria. While 70% of the 18F-NaF PET studies were performed within 30 days of the qualifying CTA, change in patient treatment after the initial CTA, such as the use of or intensity of statin therapy, might have led to a decreased in the degree of 18F-NaF uptake. However, intensification of therapy, potentially reducing plaque activity, would be expected to have led to a lessened association between PCAT density and 18F-NaF. In this study we performed the PET acquisition 1-hour after the injection of 18F-NaF limiting the number of detectable plaques with increased tracer activity (26). Delayed imaging could potentially strengthen the analysis. Further, while the PCAT density and 18F-NaF activity are correlated, the relationship between either of the measurements or their combination in predicting risk could not be assessed as our study lacked follow-up data and sufficient patient numbers to observe outcome events. Since the PCAT assessment is performed on standardly acquired CTA data, ongoing large trials assessing the role of 18F-NaF activity in predicting risk will be able to address further the complementary benefits of assessments of both 18F-NaF activity and PCAT density.
Conclusions
There is an association between noninvasive imaging markers of coronary artery plaque activity (18F-NaF uptake and pericoronary adipose tissue attenuation) and low attenuation plaque on coronary CTA in patients with clinically stable CAD and high-risk plaque features. Assessment of these two processes by means of 18F-NaF PET/CT imaging and plaque PCAT density measurements complements the CTA based plaque analysis and has potential for refining risk prediction.
Supplementary Material
Perspectives.
Competency in Medical Knowledge:
Non-invasively assessed measures of pericoronary adipose tissue related to inflammation and coronary microcalcification are associated with each other in plaques with high-risk features on coronary CTA.
Translational Outlook:
Long-term follow up in a large patient population is required to test the hypothesis that non-invasive assessment of these two biological processes translate into superior risk prediction in stable coronary artery disease patients.
Acknowledgments
Funding:
This research was supported in part by grants R01HL135557 and R01HL133616 from the National Heart, Lung, and Blood Institute/National Institute of Health (NHLBI/NIH) and by a grant from the Miriam & Sheldon G. Adelson Medical Research Foundation.
The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health. DEN (CH/09/002, RE/13/3/30183), MRD (FS/14/78/31020) and MD (FS/17/79/33226) are supported by the British Heart Foundation. DEN is the recipient of a Wellcome Trust Senior Investigator Award (WT103782AIA) and MRD a Sir Jules Thorn Award for Biomedical Research Award (2015). MCW is supported by The Chief Scientist Office of the Scottish Government Health and Social Care Directorates (PCL/17/04).
Abbreviations and acronyms
- CAD
Coronary Artery Disease
- CI
Confidence Intervals
- CTA
Coronary Computed Tomography Angiography
- HU
Hounsfield Units
- LAP
Low Attenuation Plaque
- NCP
Non-calcified Plaque Volume
- PCAT
Pericoronary Adipose Tissue
- PET
Positron Emission Tomography
- RCA
Right Coronary Artery
- SUVmax
Maximum Standard Uptake Value
- TBR
Target to Background Ratio
- 18F-NaF
18F-sodium Fluoride
References:
- 1.Virmani R, Kolodgie FD, Burke AP, Farb A, Schwartz SM. Lessons from sudden coronary death: a comprehensive morphological classification scheme for atherosclerotic lesions. Arterioscler Thromb Vasc Biol 2000;20:1262–75. [DOI] [PubMed] [Google Scholar]
- 2.Narula J, Nakano M, Virmani R, et al. Histopathologic characteristics of atherosclerotic coronary disease and implications of the findings for the invasive and noninvasive detection of vulnerable plaques. J Am Coll Cardiol 2013;61:1041–51. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Ferencik M, Mayrhofer T, Bittner D et al. Use of high-risk coronary atherosclerotic plaque detection for risk stratification of patients with stable chest pain: a secondary analysis of the PROMISE randomized clinical trial. JAMA Cardiol 2018; 3: 144–152 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Motoyama S, Ito H, Sarai M, et al. Plaque Characterization by Coronary Computed Tomography Angiography and the Likelihood of Acute Coronary Events in Mid-Term Follow-Up. J Am Coll Cardiol 2015; 66(4): 337–46. [DOI] [PubMed] [Google Scholar]
- 5.Otsuka K, Fukuda S, Tanaka A, Nakanishi K, Taguchi H, Yoshikawa J, Shimada K, Yoshiyama M. Napkin-ring sign on coronary CT angiography for the prediction of acute coronary syndrome. JACC Cardiovasc Imaging 2013;6:448–457. doi: 10.1016/j.jcmg.2012.09.016. [DOI] [PubMed] [Google Scholar]
- 6.Ferencik M, Hoffmann U. High-Risk Coronary Plaque on Computed Tomography Angiography Time to Recognize a New Imaging Risk Factor. CIRC Cardiovasc Imaging 2018;11:e007288. [DOI] [PubMed] [Google Scholar]
- 7.Joshi NV, Vesey AT, Williams MC, et al. 18F-fluoride positron emission tomography for identification of ruptured and high-risk coronary atherosclerotic plaques: a prospective clinical trial. Lancet 2014; 383(9918): 705–13. [DOI] [PubMed] [Google Scholar]
- 8.Antonopoulos AS, Sanna F, Sabharwal N, et al. Detecting human coronary inflammation by imaging perivascular fat. Science translational medicine 2017; 9(398). [DOI] [PubMed] [Google Scholar]
- 9.Marwan M Hell M Schuhback A et al. CT attenuation of pericoronary adipose tissue in normal versus atherosclerotic coronary segments as defined by intravascular ultrasound. J Comput Assist Tomogr 2017; 41: 762–767 [DOI] [PubMed] [Google Scholar]
- 10.Hedgire S Baliyan V Zucker EJ et al. Perivascular epicardial fat stranding at coronary CT angiography: a marker of acute plaque rupture and spontaneous coronary artery dissection. Radiology 2018; 287: 808–815 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Oikonomou EK, Marwan M, Desai MY et al. 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; (published online Aug 28.) [DOI] [PMC free article] [PubMed]
- 12.Goeller M, Achenbach S, Cadet S et al. Pericorary adipose tissue computed tomography attenuation and high-risk plaque characteristics in acute coronary syndrome compared with stable coronary artery disease. JAMA Cardio 2018; epub ahead of print [DOI] [PMC free article] [PubMed]
- 13.Hell MM, Motwani M, Otaki Y et al. Quantitative global plaque characteristics from coronary computed tomography angiography for the prediction of future cardiac mortality during long-term follow-up. Eur Heart J Cardiovasc Imaging 2017;18:1331–1339. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Leipsic J, Abbara S, Achenbach S, Cury R, Earls JP, Mancini GJ et al. SCCT guidelines for the interpretation and reporting of coronary CT angiography: a report of the Society of Cardiovascular Computed Tomography Guidelines Committee. J Cardiovasc Comput Tomogr 2014;8:342–58. [DOI] [PubMed] [Google Scholar]
- 15.Agatston AS, Janowitz WR, Hildner FJ, Zusmer NR, Viamonte M Jr, Detrano R. Quantification of coronary artery calcium using ultrafast computed tomography. J Am Coll Cardiol 1990;15:827–32. [DOI] [PubMed] [Google Scholar]
- 16.Doris MK, Otaki Y, Krishnan SK, et al. Optimization of reconstruction and quantification of motion-corrected coronary PET-CT. J Nucl Cardiol 2018; epub ahead of print [DOI] [PMC free article] [PubMed]
- 17.Rubeaux M, Joshi N, Dweck MR, Fletcher A, Motwani M, Thomson LE, et al. Motion correction of 18F-sodium fluoride PET for imaging coronary atherosclerotic plaques. J Nucl Med. Society of Nuclear Medicine; 2015. October 15;:jnumed.115.162990.p [DOI] [PubMed]
- 18.Aikawa E, Nahrendorf M, Figueiredo JL et al. Osteogenesis associates with inflammation in early-stage atherosclerosis evaluated by molecular imaging in vivo, Circulation 2007;116:2841e2850. [DOI] [PubMed]
- 19.New S, Goettsch C, Aikawa M et al. Macrophage-derived matrix vesicles: an alternative novel mechanism for microcalcification in atherosclerotic plaques, Circ Res 2013;113:72e77. [DOI] [PMC free article] [PubMed]
- 20.Johson RC, Leopold JA, Loscalzo J. Vascular calcification: pathobiological mechanisms and clinical implications. Circ Res 2006;99:1044–59. [DOI] [PubMed] [Google Scholar]
- 21.Irkle A, Vesey AT, Lewis DY et al. Identifying active vascular microcalcification by (18)F-sodium fluoride positron emission tomography. Nat Commun 2015;6:7495. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.McKenney-Drake ML, Territo PR, Salavati A et al. 18F-NaF PET Imaging of Early Coronary Artery Calcification. JACC Cardiovasc Imaging 2016;9:627–628 [DOI] [PubMed] [Google Scholar]
- 23.Nerlekar N, Ha FJ, Cheshire C, et al. Computed tomographic coronary angiography- derived plaque characteristics predict major adverse cardiovascular events. Circ Cardiovasc Imaging 2017;10:e006973. [DOI] [PubMed] [Google Scholar]
- 24.Detrano R, Guerci AD, Carr JJ, et al. Coronary calcium as a predictor of coronary events in four racial or ethnic groups. N Engl J Med 2008;358:1336–45. [DOI] [PubMed] [Google Scholar]
- 25.Min JK, Shaw LJ, Devereux RB, Okin PM, Weinsaft JW, Russo DJ et al. Prognostic value of multidetector coronary computed tomographic angiography for prediction of all-cause mortality. J Am Coll Cardiol 2007;50:1161–70. [DOI] [PubMed] [Google Scholar]
- 26.Kwiecinski J, Berman DS, Lee SE et al. Three-hour delayed imaging improves assessment of coronary 18F-sodium fluoride PET. J Nucl Med 2018. (published online Sep 13). [DOI] [PMC free article] [PubMed]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.