Relationship between impaired myocardial blood flow by positron emission tomography and low-attenuation plaque burden and pericoronary adipose tissue attenuation from coronary computed tomography: From the prospective PACIFIC trial

Abstract

Background.

Positron emission tomography (PET) is the clinical gold standard for quantifying myocardial blood flow (MBF). Pericoronary adipose tissue (PCAT) attenuation may detect vascular inflammation indirectly. We examined the relationship between MBF by PET and plaque burden and PCAT on coronary CT angiography (CCTA).

Methods.

This post hoc analysis of the PACIFIC trial included 208 patients with suspected coronary artery disease (CAD) who underwent > ¹⁵O]H₂O PET and CCTA. Low-attenuation plaque (LAP, < 30HU), non-calcified plaque (NCP), and PCAT attenuation were measured by CCTA.

Results.

In 582 vessels, 211 (36.3%) had impaired per-vessel hyperemic MBF (≤ 2.30 mL/min/g). In multivariable analysis, LAP burden was independently and consistently associated with impaired hyperemic MBF (P = 0.016); over NCP burden (P = 0.997). Addition of LAP burden improved predictive performance for impaired hyperemic MBF from a model with CAD severity and calcified plaque burden (P < 0.001). There was no correlation between PCAT attenuation and hyperemic MBF (r = 2 0.11), and PCAT attenuation was not associated with impaired hyperemic MBF in univariable or multivariable analysis of all vessels (P > 0.1).

Conclusion.

In patients with stable CAD, LAP burden was independently associated with impaired hyperemic MBF and a stronger predictor of impaired hyperemic MBF than NCP burden. There was no association between PCAT attenuation and hyperemic MBF. (J Nucl Cardiol 2023;30:1558–69.)

INTRODUCTION

Positron emission tomography (PET) is a clinical gold standard to quantify myocardial blood flow (MBF).¹ Impaired hyperemic MBF by PET, indicating coronary microvascular dysfunction, is strongly associated with adverse events^2,3 and showed higher predictability for significant coronary artery disease (CAD) compared to abnormal myocardial perfusion by single-photon emission computed tomography.⁴ Coronary computed tomography angiography (CCTA) can be used to assess coronary atherosclerotic plaque and surrounding tissue in addition to stenosis severity. A recent study showed that quantitative analysis of low-attenuation plaque (LAP, defined by attenuation < 30 Hounsfield units > HU]), representing necrotic core, was the only predictor for adverse events after adjustment for calcified plaque, non-calcified plaque (NCP), and obstructive CAD.⁵ However, whether those prognostic imaging biomarkers are related to hyperemic MBF by [¹⁵O]H₂O PET has not been well studied.

Pericoronary adipose tissue (PCAT) has a close bidirectional relationship with the underlying coronary artery and a simple measure of its attenuation (or density) on CCTA has been recently shown to act as a surrogate marker of coronary inflammation.^6,7 PCAT attenuation is higher in the presence of coronary inflammation, reflecting phenotypic changes in the adipocytes.⁷ Recent studies have shown that PCAT attenuation is independently associated with myocardial flow reserve (MFR) and downstream myocardial perfusion assessed by various imaging modalities.^8,9 However, Van Diemen et al. recently reported a lack of correlation between PCAT attenuation and hyperemic MBF by [¹⁵O]H₂O PET (r = 0.08).¹⁰ It is established that patients with obstructive CAD have higher epicardial adipose tissue volume,^11,12 and a negative correlation exists between PCAT attenuation and epicardial adipose tissue or PCAT volume.^13,14 Therefore, we hypothesized that PCAT attenuation is associated with hyperemic MBF depending on the CAD severity or PCAT volume. The main aim of our study was to examine the relationship between hyperemic MBF measured using > ¹⁵O]H₂O PET and quantitative plaque burden and PCAT attenuation along with CAD severity assessed on CCTA. We also sought to assess the incremental value of quantitative plaque and PCAT analyses compared to stenosis, for predicting impaired hyperemic MBF.

METHODS

Study population

This was a post hoc substudy of the PACIFIC (Prospective Comparison of Cardiac PET/CT, SPECT/CT Perfusion Imaging and CT Coronary Angiography with Invasive Coronary Angiography; NCT01521468) trial.¹⁵ In this single-center study, 208 consecutive patients with suspected stable CAD prospectively underwent CCTA, PET, and invasive coronary angiography (ICA) with fractional flow reserve (FFR) within a 2-week interval, regardless of imaging results. Patients with previously documented CAD, signs of prior myocardial infarction, atrial fibrillation, renal failure, and contraindication to adenosine were not eligible. The trial was approved by the institutional ethics committee and all participants provided written informed consent.

[¹⁵O]H₂O positron emission tomography

Patients were scanned on a hybrid PET/CT device (Philips Gemini TF64, Philips Healthcare), as previously described.¹⁵ In summary, a dynamic PET perfusion scan was performed during rest and adenosine (140 μ/kg/min) induced hyperemia, using 370 MBq of > ¹⁵O]H₂O as the radioactive tracer. A low-dose CT scan immediately followed the dynamic PET sequence to allow for attenuation correction. Reconstructed PET images were sent to a blinded core laboratory (Turku University Hospital, Turku, Finland), where resting and hyperemic quantitative MBF were calculated for all 3 major vascular territories derived from standard segmentation: left anterior descending (LAD), left circumflex (LCX), and right coronary artery (RCA). The impaired hyper-emic MBF was defined as at least 2 adjacent myocardial segments with a hyperemic flow of 2.30 mL/min/g or less.^4,15,16 Since hyperemic MBF on > ¹⁵O]H₂O PET has shown superiority for predicting functionally significant CAD beyond MFR,^16,17 we chose absolute hyperemic MBF as the reference standard.

Coronary computed tomography angiography

All patients underwent CCTA using a 256-slice CT scanner (Philips Brilliance iCT, Philips Healthcare, Best, the Netherlands). CCTA acquisition parameters were as follows: collimation 128 × 0.625 mm, gantry rotation 270 ms, tube current 200–360 mA (according to BMI), and tube voltage 120 kV. Prior to scanning, sublingual nitroglycerine was administered to all patients and metoprolol as required, aiming for a heart rate less than 65 beats/min. An intravenous bolus of 100 mL of iodinated contrast was injected at 5.7 mL/s. The scan was triggered using an automatic bolus-tracking technique with a region of interest placed in the descending thoracic aorta. Prospective electrocardiogram gating was performed at 75% of the R–R interval. CCTA data were transmitted to independent and blinded core laboratory (St Paul’s Hospital, Vancouver, British Columbia, Canada) for the assessment of diameter stenosis severity. All coronary segments ≥ 2 mm in diameter were visually graded, and CAD severity for each vessel was classified as minimal stenosis (< 10%), mild stenosis (10–49%), moderate stenosis (50–69%), and severe stenosis (70–100%).

Quantitative plaque analysis

Standardized plaque quantification was performed by an independent and blinded core laboratory (Cedars-Sinai Medical Center, Los Angeles, CA, USA) using semiautomated software (Autoplaque version 2.5, Cedars-Sinai Medical Center).^18–21 The proximal and distal limits of individual coronary lesions were defined by expert readers (K.K., D.H., and A.L.). Contouring of the vessel wall and lumen was automatic, with manual adjustment as required. Adaptive scan-specific attenuation thresholds for NCP and calcified plaque were automatically generated, as previously described.⁵ Plaque volume (mm³) was summed on a per-vessel level for the following plaque components: total plaque, calcified plaque, NCP, and LAP (defined by attenuation < 30 HU). The respective plaque burdens (%) were calculated as: plaque volume × 100% / vessel volume.²⁰

PCAT analysis

PCAT analysis was performed in all three major epicardial coronary arteries: LAD, LCX, and RCA, using semiautomated software (Autoplaque v2.5, Cedars-Sinai Medical Center). For the LAD and LCX, we analyzed a proximal 40 mm segment beginning at each coronary artery ostium. For the RCA, we analyzed a 40-mm segment beginning at 10 mm from the RCA ostium. PCAT was automatically sampled in 3D layers, moving radially outward from the vessel wall in 1 mm increments. Adipose tissue was defined as all voxels with attenuation between − 190 HU and − 30 HU.^5,16 Total volume and mean PCAT attenuation in HU were measured at a outer radial distance of 3 mm from the vessel wall (similar to mean luminal diameter of the analyzed segments).^22–24 Vessels with suboptimal image quality for PCAT assessment were excluded from PCAT analysis.

Invasive coronary angiography and FFR measurement

ICA imaging was performed according to a standard protocol. FFR was measured by using a 0.014-inch FFR guidewire introduced through a 5− or 6-F catheter. All major coronary arteries were routinely interrogated with FFR; except for in tight lesions > 90% to avoid the risk of coronary dissection by the FFR guidewire. FFR was calculated as the ratio of mean distal intracoronary pressure to mean arterial pressure during maximal hyperemia induced by intracoronary or intravenous adenosine infusion. The nonsignificant FFR was defined as FFR > 0.80.¹⁶

Statistical analysis

Categorical variables are shown as numbers and percentages and continuous variables are shown as mean ± SD or median values (interquartile range). Categorical variables were compared by the χ² test, and continuous variables were compared by the Student t test or Mann–Whitney U test, as appropriate. The Kruskal–Wallis ANOVA was used for multiple comparisons, followed by Holm post hoc test. To assess the association of plaque burdens and PCAT attenuation with impaired hyperemic MBF, a logistic regression analysis was performed, accounting for intra-patient clustering of vessels. We performed multivariable logistic regression analysis for five separate models, as follows: model 1, adjusted by CAD severity and conventional CAD risk factors including age, sex, BMI, hypertension, hyperlipidemia, and diabetes; model 2, model 1 plus total plaque burden, PCAT attenuation and volume; model 3, model 1 plus LAP burden, PCAT attenuation and volume; model 4, model 1 plus NCP burden, PCAT attenuation and volume; model 5, model 4 plus LAP burden. We adjusted for LAP and NCP burden, given that previous studies have shown their association with cardiac death and acute coronary syndrome.^20,25 Before the multivariable analyses, the multicollinearity of the independent variables was checked using the variance inflation factor, and these values were less than four.²⁶ Global χ² analyses and a likelihood ratios test were used to evaluate the incremental value of quantitative plaque and PCAT assessment. To explore the predictive value of plaque analysis for vessels with microvascular disease without functionally significant epicardial stenosis, we repeated the logistic regression analysis in vessels with non-significant FFR (> 0.80). We performed statistical analyses with JMP 14.2.0 (SAS Institute Inc, Cary, NC) and R version 3.5.3 (R Foundation for Statistical Computing, Vienna, Austria). A 2-sided P < 0.05 was considered statistically significant.

RESULTS

Study population and CCTA characteristics according to hyperemic MBF

The final study population consisted of 208 patients with 582 vessels, after exclusion of 19 vessels without measurements of hyperemic MBF and 23 vessels due to poor image quality or inadequate vessel length (< 40 mm) for analysis. Baseline patient characteristics are shown in Supplementary Table 1. Mean age of this study population was 58.1 ± 8.7 years and 132 (63.5%) were male. Vessel-based CCTA characteristics according to normal or impaired hyperemic MBF are shown in Table 1.

Table 1.

Per-vessel CCTA Characteristics according to MBF

	All vessels	Normal hyperemic MBF (> 2.30 mL/min/g)	Impaired hyperemic MBF (≤ 2.30 mL/min/g)	P-value
Number	582	371	211
Diameter stenosis
Minimal stenosis	150 (25.8)	130 (35.0)	20 (9.5)	< 0.001
Mild stenosis	228 (39.2)	163 (43.9)	65 (30.8)
Moderate stenosis	105 (18.0)	58 (15.6)	47 (22.3)
Severe stenosis	99 (17.0)	20 (5.4)	79 (37.4)
Total plaque volume (mm3)	235.4 ± 299.0	134.9 ± 187.7	427.2 ± 370.3	< 0.001
LAP volume (mm3)	32.6 ± 51.1	16.3 ± 27.9	63.8 ± 68.3	< 0.001
NCP volume (mm3)	186.3 ± 223.9	111.4 ± 151.7	329.1 ± 266.4	< 0.001
CP volume (mm3)	51.2 ± 108.1	24.6 ± 51.2	101.9 ± 158.7	< 0.001
Total plaque burden (%)	34.8 ± 24.7	28.1 ± 24.1	47.6 ± 20.4	< 0.001
LAP burden (%)	4.6 ± 4.8	3.3 ± 3.8	7.2 ± 5.3	< 0.001
NCP burden (%)	28.9 ± 21.5	23.9 ± 21.1	38.6 ± 18.8	< 0.001
CP burden (%)	6.0 ± 8.0	4.2 ± 6.3	9.3 ± 9.6	< 0.001
PCAT volume (mm3)	1171.8 ± 379.1	1116.1 ± 358.1	1269.8 ± 395.6	< 0.001
PCAT attenuation (HU)	− 85.7 ± 7.5	− 85.9 ± 7.3	− 85.4 ± 7.8	0.486

Values are expressed as n (%) or mean ± SD

CCTA, coronary computed tomography angiography; CP, calcified plaque; HU, Hounsfield unit; LAP, low-attenuation plaque; MBF, myocardial blood flow; NCP, non-calcified plaque; PCAT, pericoronary adipose tissue an independent predictor of impaired hyperemic MBF (Supplemental Table 4).

Association between hyperemic MBF, quantitative plaque analysis, PCAT attenuation, PCAT volume, and CAD severity

Figure 1 shows an example of quantitative plaque analysis in the LAD (Figure 1A and B) and polar map of hyperemic MBF (Figure 1C). This LAD had moderate stenosis and high LAP burden on CCTA, and the LAD territory showed significantly impaired hyperemic MBF by > ¹⁵O]H₂O PET (Figure 1). Figure 2A and B shows examples of PCAT attenuation maps on CCTA. Both vessels had moderate stenosis, and the vessel with high PCAT attenuation (Figure 2B) showed impaired hyperemic MBF.

Figure 1.

Case example of curved multiplanar reconstruction of the LAD showing NCP (red) and calcified plaque (yellow) (A). LAP is shown in orange on 3D reconstruction image (B). Polar map of hyperemic MBF (C). CAD, coronary artery disease; LAD, left anterior descending artery; LAP, low-attenuation plaque; MBF, myocardial blood flow; NCP, non-calcified plaque.

Figure 2.

PCAT attenuation maps on CCTA with normal MBF (A) with impaired MBF (B). The vessels in Figure 2-B and Figure 1 represent the same vessel. CAD, coronary artery disease; CCTA, coronary computed tomography angiography; HU, Hounsfield unit; LAP, low-attenuation plaque; MBF, myocardial blood flow; PCAT, pericoronary adipose tissue.

In 582 vessels, 211 (36.3%) vessels had impaired hyperemic MBF. The vessels with impaired hyperemic MBF had higher total plaque, LAP, NCP, and calcified plaque burden (all P < 0.001) (Table 1). There was a graded reduction in hyperemic MBF with increasing CAD severity (all P < 0.05) (Figure 3). There was a moderate negative correlation between hyperemic MBF and the burdens of plaque components (r = − 0.33 to − 0.41) (Figure 4). Although there was an overall weak correlation between hyperemic MBF and PCAT attenuation (r = − 0.11), a moderate negative correlation was observed between hyperemic MBF and PCAT attenuation in vessels with moderate stenosis (r = − 0.41) (Figure 5A). There was a weak negative correlation between hyperemic MBF and PCAT volume (r = − 0.21) (Figure 5B). Similar correlation was observed between hyperemic MBF and LAP burden or PCAT attenuation for each of three coronary arteries (Supplemental Figure 1).

Figure 3.

Per-vessel hyperemic MBF according to CAD severity. CAD, coronary artery disease; IQR, inter quartile range; MBF, myocardial blood flow.

Figure 4.

Correlation between hyperemic MBF and total plaque burden (A), LAP burden (B), or NCP burden (C) according to CAD severity.

Figure 5.

Correlation between hyperemic MBF and PCAT attenuation (A) or PCAT volume (B) according to CAD severity.

PCAT attenuation was negatively correlated with PCAT volume in all vessels (r = − 0.52; 95%CI − 0.58 to − 0.46; P < 0.001) (Supplemental Figure 2). Similar correlations were observed in each major epicardial coronary artery (r = − 0.55 for LAD, r = − 0.60 for LCX, and r = − 0.51 for RCA) (Supplemental Figure 3).

Association of impaired hyperemic MBF with plaque components and PCAT

In univariable analysis, CAD severity, total plaque, LAP, and NCP burden, and PCAT volume were significantly associated with impaired hyperemic MBF in all vessels, but PCAT attenuation was not (Table 2). In multivariable analysis, total plaque burden, LAP and NCP burden and PCAT volume were independent predictors of impaired hyperemic MBF next to CAD severity (model 2 to 4) (Table 2). In multivariable analysis containing LAP and NCP in the same model, LAP burden and PCAT volume were independent predictors next to CAD severity, but NCP burden was not (model 5) (Table 2).

Table 2.

Univariable and multivariable logistic regression analyses for the prediction of impaired hyperemic MBF in overall vessel

	Univariable		Multivariable (Model 1)		Multivariable (Model 2)		Multivariable (Model 3)		Multivariable (Model 4)		Multivariable (Model 5)
	OR (95% CI)	P-value	OR (95% CI)	P-value	OR (95% CI)	P-value	OR (95% CI)	P-value	OR (95% CI)	P-value	OR (95% CI)	P-value
Minimal stenosis	Reference	-	Reference	-	Reference	-	Reference	-	Reference	-	Reference	-
Mild stenosis	2.59 (1.33–5.05)	0.005	2.02 (1.01–4.04)	0.046	0.91 (0.41–2.05)	0.825	1.20 (0.58–2.49)	0.618	1.08 (0.50–2.33)	0.850	1.20 (0.56–2.60)	0.638
Moderate stenosis	5.27 (2.57–10.82)	< 0.001	4.26 (2.10–8.67)	< 0.001	2.01 (0.91–4.46)	0.086	2.47 (1.17–5.22)	0.017	2.42 (1.12–5.21)	0.024	2.47 (1.17–5.23)	0.018
Severe stenosis	25.68 (11.43–57.69)	< 0.001	17.75 (7.82–40.32)	< 0.001	5.94 (2.20–16.04)	< 0.001	6.89 (2.70–17.57)	< 0.001	7.58 (2.93–19.62)	< 0.001	6.90 (2.66–17.86)	< 0.001
Total plaque burden (%)	1.04 (1.03–1.05)	< 0.001	-	-	1.02 (1.01–1.03)	0.003	-	-	-	-	-	-
LAP burden (%)	1.23 (1.15–1.31)	< 0.001	-	-	-	-	1.12 (1.05–1.28)	< 0.001	-	-	1.12 (1.02–1.22)	0.016
NCP burden (%)	1.04 (1.02–1.05)	< 0.001	-	-	-	-	-	-	1.02 (1.00–1.03)	0.010	1.00 (0.98–1.02)	0.997
PCAT attenuation (HU)	1.04 (0.90–1.20)	0.589	-	-	1.03 (0.99–1.07)	0.170	1.03 (0.99–1.07)	0.184	1.03 (0.99–1.07)	0.187	1.03 (0.99–1.07)	0.182
PCAT volume (per 100 mm3)	1.11 (1.06–1.17)	< 0.001	-	-	1.08 (1.00–1.15)	0.038	1.07 (1.00–1.15)	0.048	1.08 (1.01–1.16)	0.022	1.07 (1.00–1.15)	0.046

Bold values indicate significant (P < 0.05)

HU, Hounsfield unit; LAP, low-attenuation plaque; MBF, myocardial blood flow; NCP, non-calcified plaque; PCAT, pericoronary adipose tissue

In all vessels, PCAT attenuation did not associate with impaired hyperemic MBF in either univariable or multivariable analysis (Table 2). In vessels with moderate stenosis, PCAT attenuation was significantly associated with impaired hyperemic MBF in both univariable and multivariable analyses (adjusted odds ratio 1.17, 95%CI 1.04–1.33; P = 0.011) (Table 3).

Table 3.

Univariable and multivariable logistic regression analyses for the prediction of impaired hyperemic MBF in each CAD severity

	Univariable		Multivariable
	OR (95% CI)	P-value	OR (95% CI)	P-value
Minimal stenosis
LAP burden (%)	1.03 (0.94–1.14)	0.505	1.08 (0.98–1.20)	0.122
PCAT attenuation (HU)	1.02 (0.94–1.11)	0.634	0.96 (0.89–1.04)	0.314
PCAT volume (per 100 mm3)	0.92 (0.80–1.06)	0.261	0.82 (0.67–1.00)	0.054
Mild stenosis
LAP burden (%)	1.11 (0.999–1.23)	0.052	1.02 (0.90–1.17)	0.715
PCAT attenuation (HU)	0.98 (0.94–1.03)	0.461	1.03 (0.98–1.09)	0.267
PCAT volume (per 100 mm3)	1.13 (1.03–1.23)	0.008	1.17 (1.05–1.30)	0.005
Moderate stenosis
LAP burden (%)	1.30 (1.15–1.48)	< 0.001	1.38 (1.17–1.63)	< 0.001
PCAT attenuation (HU)	1.10 (1.03–1.17)	0.005	1.17 (1.04–1.33)	0.011
PCAT volume (per 100 mm3)	1.06 (0.94–1.19)	0.343	1.03 (0.88–1.35)	0.424
Severe stenosis
LAP burden (%)	1.31 (1.13–1.53)	< 0.001	1.31 (1.06–1.62)	0.011
PCAT attenuation (HU)	1.00 (0.93–1.07)	0.938	0.98 (0.86–1.11)	0.733
PCAT volume (per 100 mm3)	1.16 (1.00–1.35)	0.044	1.03 (0.82–1.30)	0.787

Bold values indicate significant (P < 0.05)

HU, Hounsfield unit; LAP, low-attenuation plaque; MBF, myocardial blood flow; PCAT, pericoronary adipose tissue

We conducted subanalysis in vessel with high and low PCAT volume. High PCAT volume was defined as vessel with higher than the median PCAT volume of 1167 mm³. As a result, in vessel with high PCAT volume, PCAT attenuation was significantly associated with impaired hyperemic MBF, but not in vessel with low PCAT volume (Supplemental Table 2).

We performed an additional analysis to predict (1) impaired MFR (< 2.5); or (2) impaired hyperemic MBF using a different threshold (≤ 2.0 mL/min/g) as a sensitivity analysis. LAP burden and CAD severity were consistently associated with impaired hyperemic MBF/MFR, but PCAT volume was not (Supplemental Table 3).

Incremental value of quantitative plaque and PCAT assessment

Figure 6 shows the global χ² results for the prediction of impaired hyperemic MBF. In all vessels, the global χ² was significantly improved by the addition of LAP burden to a model with CAD severity and CP burden (global χ², 150.1 vs 132.2; P < 0.001). The global χ² was not improved by the addition of PCAT attenuation to the model including LAP burden (P = 0.214) (Figure 6A). In vessels with moderate stenosis, the global χ² was significantly improved by the addition of PCAT attenuation to a model with LAP burden and CP burden (global χ², 52.0 vs 37.3; P < 0.001) (Figure 6B).

Figure 6.

Incremental value of LAP beyond CP and CAD severity in all vessels (A) and PCAT attenuation in vessels with moderate stenosis beyond LAP and CP (B) for predicting impaired hyperemic MBF. CAD, coronary artery disease; LAP, low-attenuation plaque; MBF, myocardial blood flow; CP, calcified plaque; PCAT, pericoronary adipose tissue.

Association of impaired hyperemic MBF with plaque components in vessels with nonsignificant FFR

In a total of 582 vessels, 526 vessels (90.4%) were interrogated by invasive FFR, and 427 vessels showed nonsignificant FFR (FFR > 0.80). Of these, 90 vessels (21.1%) had impaired hyperemic MBF. In univariable analysis, CAD severity, LAP and NCP burden, and PCAT volume were associated with impaired hyperemic MBF. In multivariable analysis, only LAP burden was

DISCUSSION

In the present study, we explored the association between per-vessel plaque and PCAT measures and hyperemic MBF by > ¹⁵O]H₂O PET according to CAD severity. Our primary findings are as follows: (1) in all vessels, LAP burden was independently associated with impaired hyperemic MBF even after the adjustment including NCP burden, (2) in all vessels, the addition of LAP burden significantly improved discriminatory power for predicting vessels with impaired hyperemic MBF when added to the model with CAD severity and CP burden, and (3) there was no association between PCAT attenuation and hyperemic MBF in all vessels.

Hyperemic MBF and quantitative plaque analysis

In the present study, we assessed hyperemic MBF using > ¹⁵O]H₂O PET and used hyperemic MBF as the reference standard since this measurement has shown superiority for predicting functionally significant CAD and adverse events beyond MFR.^3,16 We showed that hyperemic MBF by > ¹⁵O]H₂O PET was negatively correlated with LAP and NCP burden. Each LAP and NCP burden was independently associated with impaired hyperemic MBF in separate model. When both parameters were included in the same model, LAP burden remained an independent predictor of impaired hyperemic MBF, while NCP burden did not (Table 2). Concordant findings have been reported in prior studies.^27,28 Importantly, due to the limit sample size or less detailed quantitative plaque analysis, those previous studies have not fully elucidated the complex interplay between LAP and MBF.^27,28 Our findings suggest, for the first time, that LAP burden, representing a marker of necrotic core, is superior to predict impaired downstream myocardial perfusion compared to NCP burden.

Hyperemic MBF and PCAT attenuation

The assessment of PCAT attenuation can potentially detect vascular inflammation indirectly. Inflamed artery releases inflammatory cytokines that prevent lipid accumulation and preadipocyte differentiation and lead changes of adipose tissue composition to less lipid content, thereby increasing PCAT attenuation.⁷ PCAT seems to function as a sensor of vascular inflammation rather than a driver.⁷ Kwiecinski et al. showed that PCAT attenuation was positively correlated with focal ¹⁸F-NaF PET uptake, possibly represents plaque activity, in patients with borderline stenosis and high-risk plaque features on CCTA.²⁹

Kanaji et al. showed that PCAT attenuation was significantly associated with impaired global MFR by phase-contrast cine-magnetic resonance imaging in patients with non-ST elevation acute coronary syndrome⁸ and Nomura et al. showed PCAT attenuation was significantly associated with impaired MFR using ⁸²Rubidium PET.⁹ Recently, Kanaji et al. showed negative moderate correlation (r = − 0.36) between PCAT attenuation and MFR in patients with single vessel intermediate CAD (30–90% stenosis),³⁰ thus high PCAT attenuation or perivascular inflammation may represents impairment of endothelial and microvascular function and play an important role in reducing hyper-emic MBF.

We observed significant association between impaired hyperemic MBF and PCAT attenuation in limited vessel (i.e., vessel with moderate stenosis or high PCAT volume), but not in overall vessels. Since PCAT attenuation and volume are negatively correlated, high PCAT attenuation in vessel with low PCAT volume may overestimate vascular inflammation or represent normal adipose tissue with small adipocyte size and low lipid content. The difference of PCAT attenuation between the case and reference is small, less than 10 HU in prior studies.^7–10,30,31 Given the potential influence of different CT scanner and acquisition and reconstruction parameters on PCAT attenuation,³² it would be challenging to utilize this inflammatory marker in clinical practice. In line with previous studies, the between-group difference in PCAT attenuation was small in the present study. It might be important to interpret PCAT attenuation by not only its HU but also the severity of CAD and PCAT volume.

Our findings are in contrast with a previous study showing no significant difference were observed in MFR between high and low PCAT attenuation in vessel with obstructive CAD (> 50% stenosis); however, PCAT volume was not assessed in the previous study.⁹ Since both studies were conducted in a single center and a relatively small sample size, further study is warranted to clarify this conflicting result. Additionally, Chatterjee et al. recently conducted substudy of the CORE320 trial showing that there was no association between PCAT attenuation and adverse cardiovascular events.³³ In the present study, PCAT attenuation was not associated with hyperemic MBF in all vessels. Traditional atherosclerosis markers such as CAD severity or plaque burden, especially LAP burden, were independently and consistently associated with hyperemic MBF. Therefore, coronary plaque assessment may have better predictive value for impaired hyperemic MBF or adverse cardio-vascular events compared with perivascular inflammation measured indirectly by PCAT attenuation.

Limitation

Our study had some limitations. The present study was performed at a single center and the data should be considered hypothesis-generating. Further study with larger sample size would be warranted to confirm our results. In addition, we cannot exclude an intrinsic selection bias and our findings may not be generalizable to other populations. Since we used the mean PCAT attenuation to estimate vascular inflammation, we might possibly underestimate vascular inflammation in obese patients given the larger adipocyte size and overestimate inflammation in lean patients.³⁴ Finally, we did not directly measure vascular inflammation; however, pre-vious studies have shown the association between histologic markers of inflammation and PCAT attenuation.⁷

NEW KNOWLEDGE GAINED

In coronary vessels, LAP burden is independently and consistently associated with impaired hyperemic MBF by > ¹⁵O]H₂O PET. The addition of LAP burden significantly improves the prediction for vessels with impaired hyperemic MBF. PCAT attenuation was not associated with impaired hyperemic MBF in the present study.

CONCLUSION

LAP burden was independently and consistently associated with impaired hyperemic MBF and a stronger predictor of impaired hyperemic MBF than NCP burden. PCAT attenuation, reflecting vascular inflammation, was not associated with hyperemic MBF in the present study.

Supplementary Material

Supplementary Information The online version contains supplementary material available at https://doi.org/10.1007/s12350–022-03194-z.

Abbreviations

CAD

Coronary artery disease

CCTA

Coronary computed tomography angiography

LAP

Low-attenuation plaque

MBF

Myocardial blood flow

NCP

Non-calcified plaque

PCAT

Pericoronary adipose tissue

PET

Positron emission tomography