Abstract
Aims
We aimed to investigate whether quantitative plaque features measured from coronary CT angiography (CCTA) predict ischemia by myocardial perfusion SPECT imaging (MPI).
Methods and Results
Hundred and eighty-four consecutive patients (63% males) with suspected-coronary artery disease, undergoing hybrid CCTA, and attenuation corrected solid state 99mTc stress/rest MPI and single vessel ischemia were considered. Quantitative analysis of CCTA derived non-calcified plaque (NCP), low-density NCP [< 30 Hounsfield Units (HU)] (LDNCP), calcified and total plaque burdens (%, normalized to vessel volume), maximum diameter stenosis and contrast density difference (CD, maximum difference in HU/lumen area within lesion). Normal thresholds for plaque features were defined as 95th percentile thresholds, from 40% of vessels with non-ischemic MPI regions. These vessels were excluded from further analysis. Regional ischemia (≥ 2%) was quantified from MPI. All plaque features were higher in arteries corresponding to ischemia (P < 0.003 for all). In multi-variable analysis, abnormal NCP burden [odds ratio (OR) 2.6], LDNCP burden (OR 3.9), and CD (OR 2.7) were significantly associated with ischemia, whereas stenosis ≥ 50% was not (P = 0.14). In a subset of vessels with ≥ 50% stenosis, LDNCP burden (OR 4.3, P = 0.008) and CD (OR 3.7, P = 0.029) were associated with ischemia. In subsets of vessels with stenosis 30–69% and ≥ 70%, abnormal LDNCP burden (OR 6.4, P = 0.006) and CD (OR 7.3, P = 0.02) were associated with ischemia.
Conclusions
Quantitative plaque features obtained from CCTA, LDNCP, and CD, are associated with ischemia by MPI independent of stenosis. LDNCP burden and CD are associated with ischemia in stenosis 30–69% and ≥ 70%, respectively.
Keywords: coronary plaque , ischemia , CCTA , low density non-calcified plaque , contrast density difference
Introduction
Coronary CT angiography (CCTA) can provide accurate information about anatomic extent of coronary artery disease (CAD) with consistently demonstrated high negative predictive value.1 Nevertheless, coronary plaque findings in CCTA, particularly in vessels with non-severe stenosis usually have a less clear clinical significance. Whether a coronary lesion with non-severe stenosis identified in CCTA is haemodynamically significant, is a commonly asked question in everyday clinical practice and the functional myocardial perfusion evaluation is frequently required in this scenario.
Single photon emission computed tomography (SPECT) myocardial perfusion imaging (MPI) is widely used for the detection and quantification of myocardial ischemia and has been shown to be a strong prognostic indicator of the risk for adverse cardiac events.2,3
Prior studies have shown that quantitative assessment of plaque characteristics could increase the performance and specificity of CCTA in detection of ischemia.4 Since CCTA provides information about a variety of plaque features in addition to stenosis, we aimed to investigate whether automated quantitative plaque features measured from CCTA predict ischemia by MPI, for varying stenosis severity.
Methods
Study population
Consecutive patients undergoing hybrid CCTA and stress/rest MPI scans on a Cadmium Zinc Telluride (CZT) camera for assessment of known or suspected CAD between 06/2010 and 02/2013 were included. Exclusion criteria were history of coronary artery bypass graft surgery, stent placement or myocardial infarction, coronary artery calcium (CAC) ≥ 1000 Agatston Units (AU) and multi-vessel disease as judged by visual assessment (stenosis ≥ 70% in more than one coronary artery). The study was approved by the Institutional Review Board.
CCTA
CCTA image acquisition was performed on a 64-slice CT system (GE Lightspeed VCT or Discovery HD 750, GE Healthcare, Milwaukee, WI) as previously published.5 Patients underwent an unenhanced prospectively ECG-triggered low-dose sequential CT scan of the heart for CAC scoring and a contrast-enhanced prospective ECG-triggered CT scan during inspiration breath hold as previously reported.5 Metoprolol was administrated intravenously before the examination if heart rate was higher than 65 beat/min and 2.5 mg of isosorbide dinitrate was administrated sublingually to obtain optimal image quality. Iodixanol (Visipaque 320; 320 mg/mL; GE Healthcare, Buckinghamshire, UK) was injected into an antecubital vein followed by 50 mL of saline solution via an 18-gauge catheter. Contrast media volume (40–105 mL) and flow rate (3.5–5mL/s) were adapted to body surface area.6
CCTA quantitative analysis
For each abnormal coronary segment, the following features were quantified using semi-automated Autoplaque software (Cedars-Sinai Medical Center, Los Angeles, CA). This software allows quantitative 3D assessments of plaque characteristics from CCTA (including plaque volume burden, distribution, and composition) and coronary artery remodeling.7 Coronary segments for analysis are defined by the user by setting proximal and distal points for quantification of plaque characteristics. The absolute volumes for total plaque, non-calcified plaque (NCP), and calcified plaque were computed using scan-specific thresholds.7,8 Low-density NCP (LDNCP) was defined as NCP below a preset low-density threshold (30 HU).9 The plaque burden, defined as the plaque volume normalized to the vessel volume in the plaque region (plaque volume × 100%/vessel volume), for each of these plaque components was calculated. Maximum diameter stenosis was calculated by dividing the narrowest lumen diameter by the mean of two normal non-diseased reference points. Remodelling index was determined as the ratio of maximum vessel area to that at the proximal normal reference point.10 Lesion length (in mm) was the length of the diseased vessel as computed in Autoplaq. Contrast density difference over the lesion, was computed as follows: the luminal contrast density (defined as attenuation per unit area), similar to ‘area gradient’,11 was automatically computed over 1 mm cross-sections of the arterial segment, thus it is normalized to the vessel area. The contrast density difference was defined as the maximum percent difference in contrast densities, with respect to the proximal reference cross-section (with no disease). The derivation of quantitative plaque features is shown in Figure 1. To quantify each lesion, CCTA images were examined by an experienced reader, blinded to SPECT findings. Subsequent plaque quantification was automated. Volumes of each segment were then summed to obtain total per-vessel plaque volumes. For each artery, maximum diameter stenosis, remodelling index, and contrast density difference values were reported. Final result of quantification was edited if needed by the experienced reader. Time for evaluation was approximately 3–5 min per-vessel, depending on the severity of CAD.
Figure 1.
Quantitative plaque features. Panel A shows software view for quantification of absolute volumes for non-calcified plaque (red overlay) and calcified plaque (yellow overlay). Plaque burden for these components is then calculated, defined as the plaque volume normalized to the vessel volume (green overlay). Panel B illustrates contrast density difference measure, defined as the maximum percent difference in luminal contrast density (small circle) within the lesion (segment between yellow arrows) with respect to the proximal reference cross-section (large circle).
Normal thresholds for quantitative CCTA plaque features
Normal thresholds were determined from a separate derivation group. From the vessels corresponding to non-ischemic territories, a subset of 40% of the non-ischemic vessels were removed and used as the derivation group. These vessels were used only to obtain normal thresholds for plaque features defined as >95th percentile and were excluded from further analysis exploring the relationship of plaque features and ischemia (Figure 2).12
Figure 2.
Distribution of vessels included in derivation cohort and analysis population.
MPI
For all patients, MPI was performed after CCTA on the same day. The protocol consisted in a 1-day stress/rest MPI with either: (i) standard adenosine infusion (0.14 mg/kg/min over 6 min), (ii) dobutamine infusion (incrementally administered, starting at 5 μg/kg per min and increasing at 1-min intervals to a maximal dose of 60 μg/kg per min until 85% of the age-predicted heart rate had been achieved), or (iii) bicycle stress. Out of 184 patients, 146 underwent adenosine stress, 13 underwent dobutamine tests, and 25 underwent treadmill tests. For the treadmill tests, 20 patients reached 85% of age-predicted rate, and remaining 5 reached ≥80% of age predicted rate. Approximately 60 min after injection of 99mTc-tetrofosmin, stress images were acquired over 5 min on the MPI system with a multi-pinhole collimator and 19 stationary CZT detectors.13 Rest scans were obtained with the identical protocol several minutes after administration of a three times higher dose of 99mTc-tetrofosmin.14 A low-dose 64-slice CT scan with prospective triggering was performed for attenuation correction (AC) MPI.15 CZT images were reconstructed by an iterative algorithm.
MPI quantitative analysis
The transaxial images were first segmented by QPS/QGS (Cedars QGS/QPS; Cedars-Sinai Medical Center). All image contours were reviewed by an experienced technologist. QPS software computed total perfusion deficit (TPD) independently for stress and rest scan in each vessel territory. Ischemic TPD (ITPD) was defined as the difference between stress and rest TPD for each vessel territory. The presence of regional ischemia was defined as ≥ 2% regional ITPD.16
Statistical analysis
All analysis to assess the relationship between quantitative plaque features by CCTA and ischemia by SPECT was performed per vessel, using specific vessel quantitative plaque features and corresponding regional ITPD. We also performed a sub-analysis to assess the relationship of quantitative plaque features and ischemia in different categories of stenosis grade: (i) diameter stenosis ≥ 50%, (ii) diameter stenosis 30–69%, (iii) diameter stenosis ≥ 70%,17 as well as gender-specific analysis to investigate potential gender differences in plaque characteristics associated with ischemia.
Data were tested for normal distribution by the Shapiro–Wilk test and presented as mean ± standard deviation (SD). Continuous variables were examined using Student’s t-test. Categorical variables were expressed as percentages and examined using Chi-square or Fisher’s exact test as appropriate. Logistic regression analysis was used to examine the relation between plaque features and ischemia by MPI. To examine the added improvement in predicting ischemia due to quantitative plaque features over stenosis alone, we utilized integrated discrimination improvement (IDI) analysis,18 and receiver operating characteristics (ROC) analysis. Statistical analysis was performed using STATA version 11, and Analyse-it software. IDI analyses were performed using SAS 9.2. All tests of significance were two-sided; a probability of <0.05 was considered significant.
Results
Patient characteristics
Clinical patient characteristics are listed in Table 1. Hundred and eighty-four subjects (116, 63% males) were included. Among them, 27 subjects (15%) underwent an invasive coronary angiogram after CCTA and MPI and 18 subjects underwent coronary revascularization. All coronary segments in 544 vessels were evaluable except for the 8 vessels excluded due to severe motion artifacts. Ischemia was found in territories corresponding to 71 vessels (13% of the 544 vessels), belonging to 71 patients (since patients with only single vessel disease were included). Thresholds associated in the non-ischemic derivation cohort (188 vessels) were calculated for diameter stenosis, remodelling index, lesion length, and contrast density difference (Table 2). A separate analysis group of 356 vessels (71 ischemic, 285 non-ischemic) was subsequently analysed. There were no significant differences in age, risk factors, and clinical presentation between the vessels used as derivation cohort and the vessels corresponding to non-ischemic territories included in the analysis cohort.
Table 1.
Clinical characteristics
| Parameter | All | Men | Women | |
|---|---|---|---|---|
| Number | 184 | 116 | 68 | |
| Age, years | 61 ± 9 | 60 ± 9 | 62 ± 9 | |
| BMI, kg/m2 | 28 ± 5 | 28 ± 4 | 27 ± 6 | |
| Hypertension, (%) | 61 | 61 | 62 | |
| Hyperlipidemia, (%) | 56 | 55 | 57 | |
| Diabetes, (%) | 15 | 17 | 12 | |
| Symptomatic, (%)a | 66 | 59 | 78 | |
| Typical chest pain, (%) | 16 | 18 | 12 | |
| Atypical chest pain, (%) | 41 | 32 | 57 | |
| Shortness of breath, (%) | 15 | 13 | 18 | |
Results are expressed in mean ± standard deviation or %.
BMI, body mass index.
Includes typical, atypical chest pain, and shortness of breath
Table 2.
Normal thresholds for quantitative CCTA plaque features corresponding to 95th percentile value in the derivation data
| Parameter | 95th percentile |
|---|---|
| Diameter stenosis | ≥63% |
| Total plaque volume | 296 mm3 |
| NCP volume | 267 mm3 |
| LDNCP volume | 39 mm3 |
| CP volume | 52 mm3 |
| Total plaque burden | ≥41% |
| NCP burden | ≥39% |
| LDNCP burden | ≥7% |
| Calcified plaque burden | ≥8% |
| Remodelling index | ≥1.7 |
| Lesion length | ≥56 mm |
| Contrast density difference | ≥33% |
NCP, non-calcified plaque; LDNCP, low density non-calcified plaque; CP, calcified plaque
Quantitative plaque features in vessels corresponding to ischemia
All quantitative CCTA plaque measures were higher in the arteries corresponding to the ischemic regions than in the arteries without corresponding ischemia (Table 3). Consistently, the prevalence of abnormal total plaque burden, NCP burden, LDNCP burden and contrast density difference was higher in the arteries corresponding to the ischemic territories (Figure 3). The prevalence of abnormal calcified plaque burden was not higher in the ischemic territories (6% vs. 11%, P = 0.12). The plaque features that showed higher prevalence in vessels corresponding to ischemia were examined with multi-variable analysis, adjusted by age and gender. Abnormal NCP burden, LDNCP burden, and contrast density difference were associated with the presence of ischemia. Stenosis ≥ 50%, abnormal remodelling index, and abnormal lesion length were not significant. Since NCP burden and LDNCP burden are highly correlated measures, only LDNCP burden and contrast density drop were included in the logistic regression model. Odds ratios of quantitative plaque features associated with the presence of ischemia by SPECT, adjusted by age and gender in all vessels and in different stenosis categories, are presented in Table 4.
Table 3.
Univariate analysis of quantitative CCTA plaque features in ischemic and non-ischemic vessels for the analysis cohort
| Parameter | Ischemic 71 vessels | Non-ischemic 285 vessels | P | |
|---|---|---|---|---|
| Diameter stenosis (%) | 44 ± 33 | 23 ± 27 | <0.001 | |
| Plaque volume (mm3) | Total | 114 ± 118 | 62 ± 89 | <0.001 |
| NCP | 99 ± 90 | 55 ± 80 | <0.001 | |
| LDNCP | 20 ± 24.3 | 8 ± 14 | <0.001 | |
| CP | 16 ± 29.1 | 7 ± 15 | <0.001 | |
| Plaque burden (%) | Total | 26 ± 18.4 | 15 ± 16 | <0.001 |
| NCP | 23 ± 17 | 13 ± 15 | <0.001 | |
| LDNCP | 4 ± 4 | 2 ± 1 | <0.001 | |
| CP | 3 ± 5 | 2 ± 3 | 0.002 | |
| Remodelling index | 1.1 ± 0.7 | 0.7 ± 0.7 | <0.001 | |
| Lesion length, mm | 26 ± 23 | 16 ± 20 | <0.001 | |
| Contrast density difference (%) | 27 ± 32 | 10 ± 15 | <0.001 |
NCP, non-calcified plaque; LDNCP, low-density non-calcified plaque; CP, calcified plaque; bold font signifies P < 0.05.
Figure 3.
Prevalence of abnormal CCTA plaque features in ischemic vs. non-ischemic regions for different categories of stenosis: stenosis ≥ 50% (left), stenosis 30–69% (middle), and stenosis ≥ 70% (right). Plaque features shown: NCP, non-calcified plaque burden; LDNCP, low-density non-calcified plaque burden; CD, contrast density difference. *Signifies P < 0.05 ischemic vs. non-ischemic.
Table 4.
Odds ratios of quantitative plaque features associated with the ischemia by MPI adjusted by age and gender in all vessels and in different stenosis categories
| OR | 95% CI | P | |
|---|---|---|---|
| All vessels | |||
| Stenosis ≥ 50% | 1.7 | 0.8–3.6 | 0.14 |
| Abnormal LDNCP burden | 3.9 | 1.5–9.8 | 0.004 |
| Abnormal contrast density difference | 2.7 | 1–7.4 | 0.047 |
| Vessels with stenosis ≥ 50% | |||
| Severe stenosis ≥ 70% | 0.7 | 0.2–2.4 | 0.61 |
| Abnormal LDNCP burden | 3.4 | 1.1–10.5 | 0.03 |
| Abnormal contrast density difference | 3.2 | 0.9–10.9 | 0.06 |
| Vessels with stenosis 30–69% | |||
| Significant stenosis ≥ 50% | 1.4 | 0.5–3.5 | 0.52 |
| Abnormal LDNCP burden | 6.0 | 1.5–24.1 | 0.01 |
| Abnormal contrast density difference | 1.3 | 0.2–9.5 | 0.72 |
| Vessels with stenosis ≥ 70% | |||
| Abnormal LDNCP burden | 3.2 | 0.6–15.9 | 0.15 |
| Abnormal contrast density difference | 6.3 | 1.2–33.2 | 0.03 |
OR, odds ratio; CI, confidence interval; LDNCP, low-density non-calcified plaque; bold font signifies P < 0.05.
Plaque features associated with ischemia in different stenosis categories
In the vessels with stenosis ≥ 50% (86 vessels), abnormal LDNCP burden and contrast density difference were more prevalent in the territories corresponding to ischemia, and both features were associated with ischemia in multi-variable analysis adjusted by stenosis, age, and gender. In the vessels with stenosis 30–69% (112 vessels), LDNCP burden was associated with ischemia in multi-variable analysis adjusted by stenosis, age, and gender. In the vessels with stenosis ≥ 70% (40 vessels), abnormal contrast density difference was associated with ischemia in multi-variable analysis adjusted by stenosis, age, and gender.
The prevalence of abnormal quantitative plaque features in arteries associated with ischemic and non-ischemic territories subsets for different stenosis categories are shown in Figure 3. Multi-variable analysis is shown in Table 4.
Plaque features associated with ischemia in men and women
The prevalence of abnormal quantitative plaque features in arteries associated with ischemic and non-ischemic territories in men and women are shown in Figure 4. In multi-variable analysis adjusted by age, gender, and stenosis ≥ 50%, abnormal LDNCP burden was associated with ischemia in men. In women, abnormal contrast density difference was associated with ischemia; however abnormal LDNCP burden did not reach significance (P = 0.06) (Table 5). An example of a case with abnormal plaque features and corresponding ischemia is shown in Figure 5.
Figure 4.
Prevalence of abnormal CCTA plaque features in ischemic vs. non-ischemic regions for all vessels (left), men (middle) and women (right). Plaque features shown: NCP, non-calcified plaque burden; LDNCP, low-density non-calcified plaque burden; CD, contrast density difference. *Signifies P < 0.05 ischemic vs. non-ischemic.
Table 5.
Odds ratios of quantitative plaque features associated with the presence of ischemia by MPI adjusted by age in sub-analysis in men and women
| OR | 95% CI | P | |
|---|---|---|---|
| Men, 194 vessels | |||
| Stenosis ≥ 50% | 2.1 | 0.9–5.1 | 0.08 |
| Abnormal LDNCP burden | 4.0 | 1.3–12.3 | 0.01 |
| Abnormal contrast density difference | 1.7 | 0.5–5.6 | 0.4 |
| Women, 162 vessels | |||
| Stenosis ≥ 50% | 0.9 | 0.18–4 | 0.8 |
| Abnormal LDNCP burden | 6.2 | 0.9–42.4 | 0.06 |
| Abnormal contrast density difference | 8.1 | 1.3–49.7 | 0.02 |
OR, odds ratio; CI, confidence interval; LDNCP, low-density non-calcified plaque; bold font signifies P < 0.05.
Figure 5.
CCTA and MPI results in a 57-year-old symptomatic man. In Panel A, CCTA straightened and cross-sectional views of proximal LAD artery with plaque overlay (red: non-calcified and yellow: calcified). Diameter stenosis was ≥ 70%. In Panel B, corresponding MPI shows perfusion defect in LAD territory (white arrows) on short axis and vertical long axis images and polar maps. Ischemic TPD was 5% in the LAD territory.
Quantitative plaque features and identification of ischemia
By IDI analyses, the addition of these quantitative plaque features significantly improved discrimination of ischemia in all vessels, as well as in vessels with stenosis ≥ 50% and 30–69%. In men, the discrimination was also significantly improved, whereas in women, improvement showed a trend for significance (Table 6).
Table 6.
IDI when quantitative plaque features (low-density non-calcified plaque burden and contrast density difference) are added to stenosis grade
| IDI | 95% CI | P | |
|---|---|---|---|
| All vessels | 0.05 | 0.02–0.08 | 0.003 |
| Vessels with stenosis ≥ 50% | 0.08 | 0.02–0.15 | 0.006 |
| Vessels with stenosis 30–69% | 0.07 | 0.01–0.13 | 0.03 |
| Men | 0.05 | 0.01–0.09 | 0.01 |
| Women | 0.08 | 0.01–0.16 | 0.05 |
IDI, Integrated discrimination; CI, confidence Interval improvement; bold font signifies P < 0.05.
For all vessels, the ROC area under the curve for identifying ischemia for quantitative plaque features (low-density NCP burden and contrast density difference, when adjusted for age and gender, as standard) was 0.76 (95% CI: 0.66–0.87), significantly higher than that for maximal stenosis grade [0.60 (95% CI: 0.46–0.73)], P = 0.03.
Discussion
This study of the relationship between quantitative plaque features and the presence of ischemia identified by quantitative analysis of MPI resulted in several noteworthy findings. First, we derived the 95th percentile normal thresholds for the quantitative plaque features obtained from CCTA by automated software in arteries without corresponding ischemia. Second, we established that LDNCP burden and contrast density difference are associated with the presence of ischemia, independent of stenosis. Third, we found that the addition of these quantitative CCTA plaque features significantly improved the determination of ischemia. In a sub-analysis, we also found that LDNCP burden is the most relevant plaque feature associated with ischemia in vessels with stenosis 30–69%, while contrast density difference is the most relevant plaque feature associated with ischemia in vessels with stenosis ≥ 70%.
Certain aspects of the present work should be highlighted. The population in this study consisted of consecutive patients undergoing both CCTA and MPI on the same day; results from both tests are independently assessed. Our cohort is a low-risk group resembling a real-world population and referring criteria, unlike populations studied in previous reports evaluating associations of coronary plaques with ischemia, particularly when the defined outcome was the invasive testing. To our knowledge, this is the largest study on the association of characterized, quantitative plaque measures by CCTA with quantitative measures of ischemia by MPI. Importantly, this is the first study establishing normal thresholds for quantitative plaque measures by CCTA, above which there is a potential risk for ischemia.
The mismatch between CCTA findings and the presence of ischemia is well-known. Schuijf et al.19 reported that among patients with stenosis ≥ 50% by CCTA, only 50% had ischemia by SPECT and 15% of the patients with no stenosis ≥ 50% had perfusion abnormalities. This mismatch occurs also when MPI is compared with invasive coronary angiography. The FAME study20 reported that 65% of lesions considered to be of 50–69% stenosis were not ischemic by invasive fractional flow reserve (FFR), which is currently accepted as the invasive standard for assessment of lesion-specific ischemia.
Prior studies have shown plaque characteristics assessed from CCTA can increase the value of CCTA to predict lesion-specific ischemia.21,22 Shmilovich et al.23 examined the relationship of adverse plaque features defined visually (low attenuation plaque, positive remodelling, spotty calcifications) with MPI ischemia in 49 patients and found that plaques with both low attenuation and positive remodelling were likely to be associated with ischemia. Diaz-Zamudio et al.24 examined the relationship between quantitative plaque characteristics, and lesion-specific haemodynamic significance by invasive FFR in 56 patients. Automatic quantification of total, NCP, and LDNCP burden substantially improved determination of ischemia. In a large multi-centre study of 254 patients undergoing CCTA before invasive FFR, LDNCP volume also significantly predicted ischemic lesions by FFR.22 Compared to FFR, MPI results provide less direct estimate of ischemia and may also include ischemia on the microvascular level. However, due to its non-invasive nature MPI allows us to evaluate a low-to-intermediate risk group of patients, as opposed to other invasive methods to assess ischemia, such as FFR.
Our findings are in accordance to these previous results and build on the concept that LDNCP component adds in the discrimination of ischemia. Plaques with lipid-rich necrotic cores are known as one of the main causes of myocardial infarction and sudden cardiac death.25 Our finding that LDNCP (a surrogate for the necrotic core in CCTA) is significantly associated with myocardial ischemia suggests that lipid-rich plaques and their subsequent rupture may be the mechanistic link behind poorer clinical outcomes in patients with ischemia; however, this needs to be confirmed with follow-up studies.
Our study further presents insight about how different quantitative plaque features measured from CCTA alone such as LDNCP and a novel marker—contrast density difference—can be used. The finding that LDNCP provides incremental discrimination of ischemia beyond stenosis severity is in accordance with previous studies.21,22 In stenosis ≥ 70%, as seen in our sub-analysis, a most severe luminal narrowing might overcome the relevance of plaque composition and an indirect measure of haemodynamic significance, such as the contrast density difference,26 could be the most important feature to aid in ischemia detection.
Interestingly, in multivariable analysis, we found that the contrast density difference was the most relevant quantitative plaque measure associated with ischemia in women, over LDNCP burden, which did not reach statistical significance. While gender-specific analysis of quantitative plaque features in relation to ischemia has not been previously reported, the prospective, multi-centre PROSPECT study performed quantitative coronary angiography and intravascular ultrasound in 697 patients (24% were women) with acute coronary syndromes,27 and found that women were characterized by significantly fewer plaque ruptures, and less necrotic core volume. The authors concluded that other high-risk plaque characteristics such as smaller minimal lumen area and smaller total fibrous volume, may explain the similar cardiovascular event rates to men. Contrast density difference, by definition, is the difference in lumen attenuation per unit area over a segment, and is closely related to both minimum luminal area and change in luminal attenuation. While women have smaller coronary artery diameters, the contrast density difference tries to minimize variation in vessel sizes since it is normalized by the vessel area and also measured within one segment results finding of contrast density difference being more relevant than low density plaque burden in women might be in line with these previously reported gender-specific differences. Further studies are needed to confirm these findings.
The potential clinical applications of plaque characterization in guiding the revascularization therapy have been consistently suggested in the literature. The current limitation for its use in daily practice is the requirement of time-consuming manual methods. In this work, we show the results obtained by a relatively fast quantification tool, thus expanding its clinical potential. Further prospective, larger studies are required to confirm applicability and its impact in clinical decision.
Our study had some limitations. The enrolment of our study was retrospective and an overall referral bias could be present with patients undergoing the hybrid SPECT/CT imaging. We only considered patients with single vessel disease and did not evaluate patients with multi-vessel disease and consequently did not perform per-patient analysis; however, our aim was to evaluate the direct relationships between specific plaque features by CTA and regional ischemia by MPI. Single vessel CAD was chosen in this study to isolate the specific plaque characteristics and evaluate their effect with respect to the regional ischemia. This allowed us to avoid the effects of balanced ischemia and imprecise definition of vascular regions in MPI. Further studies analysing global plaque burden link with global ischemia by MPI in overall population could be performed. We acknowledge that the sub-analysis in women can potentially be under powered due to the number of women in this cohort (37%). Similarly, sub-analysis in the 30–69% and ≥70% stenosis groups can be underpowered and larger studies are needed to support these findings.
In conclusion, our study shows that selected quantitative plaque features from CCTA are associated with ischemia by MPI, independent of stenosis. LDNCP burden is the most relevant plaque feature associated with ischemia in vessels with stenosis 30–69%, while contrast density difference is the most relevant plaque feature associated with ischemia in vessels with stenosis ≥ 70%.
Acknowledgements
This research was supported in part by Grant R01HL089765 from the National Heart, Lung, and Blood Institute/National Institutes of Health (NHLBI/NIH) (PI: Piotr Slomka). Its contents are solely the responsibility of the authors and do not necessarily represent the official views of the NHLBI/NIH.
Conflict of interest: Cedars-Sinai Medical Center received royalties for the quantitative assessment of function, perfusion, viability by MPI, and CT plaque characteristics, a portion of which is distributed to some of the authors of this manuscript (D.B., G.G., P.S., D.D.).
Funding
This research was supported in part by grant R0HL089765 from the National Heart, Lung, and Blood Institute/National Institutes of Health (NHLBI/NIH) (PI: Piotr Slomka). Its contents are solely the responsibility of the authors and do not necessarily represent the official views of the NHLBI.
References
- 1. Mowatt G, Cook JA, Hillis GS, Walker S, Fraser C, Jia X. et al. 64-Slice computed tomography angiography in the diagnosis and assessment of coronary artery disease: systematic review and meta-analysis. Heart 2008;94:1386–93. [DOI] [PubMed] [Google Scholar]
- 2. Hachamovitch R, Berman DS, Shaw LJ, Kiat H, Cohen I, Cabico JA. et al. Incremental prognostic value of myocardial perfusion single photon emission computed tomography for the prediction of cardiac death: differential stratification for risk of cardiac death and myocardial infarction. Circulation 1998;97:535–43. [DOI] [PubMed] [Google Scholar]
- 3. Hachamovitch R, Hayes SW, Friedman JD, Cohen I, Berman DS.. Comparison of the short-term survival benefit associated with revascularization compared with medical therapy in patients with no prior coronary artery disease undergoing stress myocardial perfusion single photon emission computed tomography. Circulation 2003;107:2900–7. [DOI] [PubMed] [Google Scholar]
- 4. Rossi A, Papadopoulou SL, Pugliese F.. Quantitative computed tomographic coronary angiography: does it predict functionally significant coronary stenoses?. Circ Cardiovasc Imaging 2014;7:43–51. [DOI] [PubMed] [Google Scholar]
- 5. Husmann L, Valenta I, Gaemperli O, Adda O, Treyer V, Wyss CA. et al. Feasibility of low-dose coronary CT angiography: first experience with prospective ECG-gating. Eur Heart J 2008;29:191–7. [DOI] [PubMed] [Google Scholar]
- 6. Pazhenkottil AP, Husmann L, Buechel RR, Herzog BA, Nkoulou R, Burger IA. et al. Validation of a new contrast material protocol adapted to body surface area for optimized low-dose CT coronary angiography with prospective ECG-triggering. Int J Cardiovasc Imaging 2010;26:591–7. [DOI] [PubMed] [Google Scholar]
- 7. Dey D, Schepis T, Marwan M, Slomka PJ, Berman DS, Achenbach S.. Automated three-dimensional quantification of noncalcified coronary plaque from coronary CT angiography: comparison with intravascular US. Radiology 2010;257:516–22. [DOI] [PubMed] [Google Scholar]
- 8. Dey D, Cheng VY, Slomka PJ, Nakazato R, Ramesh A, Gurudevan S. et al. Automated 3-dimensional quantification of noncalcified and calcified coronary plaque from coronary CT angiography. J Cardiovasc Comput Tomogr 2009;3:372–82. [DOI] [PubMed] [Google Scholar]
- 9. Motoyama S, Sarai M, Harigaya H, Anno H, Inoue K, Hara T. et al. Computed tomographic angiography characteristics of atherosclerotic plaques subsequently resulting in acute coronary syndrome. J Am Coll Cardiol 2009;54:49–57. [DOI] [PubMed] [Google Scholar]
- 10. Achenbach S, Ropers D, Hoffmann U, MacNeill B, Baum U, Pohle K. et al. Assessment of coronary remodeling in stenotic and nonstenotic coronary atherosclerotic lesions by multidetector spiral computed tomography. J Am Coll Cardiol 2004;43:842–7. [DOI] [PubMed] [Google Scholar]
- 11. Steigner ML, Mitsouras D, Whitmore AG, Otero HJ, Wang C, Buckley O. et al. Iodinated contrast opacification gradients in normal coronary arteries imaged with prospectively ECG-gated single heart beat 320-detector row computed tomography. Circ Cardiovasc Imaging 2010;3:179–86. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Poulsen OM, Holst E, Christensen JM.. Calculation and application of coverage intervals for biological reference values. Pure Appl Chem 1997;69:1601–11. [Google Scholar]
- 13. Herzog BA, Buechel RR, Katz R, Brueckner M, Husmann L, Burger IA. et al. Nuclear myocardial perfusion imaging with a cadmium-zinc-telluride detector technique: optimized protocol for scan time reduction. J Nucl Med 2010;51:46–51. [DOI] [PubMed] [Google Scholar]
- 14. Buechel RR, Herzog BA, Husmann L, Burger IA, Pazhenkottil AP, Treyer V. et al. Ultrafast nuclear myocardial perfusion imaging on a new gamma camera with semiconductor detector technique: first clinical validation. Eur J Nucl Med Mol Imaging 2010;37:773–8. [DOI] [PubMed] [Google Scholar]
- 15. Herzog BA, Buechel RR, Husmann L, Pazhenkottil AP, Burger IA, Wolfrum M. et al. Validation of CT attenuation correction for high-speed myocardial perfusion imaging using a novel cadmium-zinc-telluride detector technique. J Nucl Med 2010;51:1539–44. [DOI] [PubMed] [Google Scholar]
- 16. Slomka PJ, Nishina H, Berman DS, Akincioglu C, Abidov A, Friedman JD. et al. Automated quantification of myocardial perfusion SPECT using simplified normal limits. J Nucl Cardiol 2005;12:66–77. [DOI] [PubMed] [Google Scholar]
- 17. Abbara S, Arbab-Zadeh A, Callister TQ, Desai MY, Mamuya W, Thomson L. et al. SCCT guidelines for performance of coronary computed tomographic angiography: a report of the society of cardiovascular computed tomography guidelines committee. J Cardiovasc Comput Tomogr 2009;3:190–204. [DOI] [PubMed] [Google Scholar]
- 18. Pencina MJ, D’Agostino RB, Pencina KM, Janssens AC, Greenland P.. Interpreting incremental value of markers added to risk prediction models. Am J Epidemiol. 2012;176:473–81. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Schuijf JD, Wijns W, Jukema JW, Atsma DE, de Roos A, Lamb HJ. et al. Relationship between noninvasive coronary angiography with multi-slice computed tomography and myocardial perfusion imaging. J Am Coll Cardiol 2006;48:2508–14. [DOI] [PubMed] [Google Scholar]
- 20. Tonino PA, Fearon WF, De Bruyne B, Oldroyd KG, Leesar MA, Ver Lee PN. et al. Angiographic versus functional severity of coronary artery stenoses in the FAME study fractional flow reserve versus angiography in multivessel evaluation. J Am Coll Cardiol 2010;55:2816–21. [DOI] [PubMed] [Google Scholar]
- 21. Park HB, Heo R, o Hartaigh B, Cho I, Gransar H, Nakazato R. et al. Atherosclerotic plaque characteristics by CT angiography identify coronary lesions that cause ischemia: a direct comparison to fractional flow reserve. JACC Cardiovasc Imaging 2015;8:1–10. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Gaur S, Ovrehus KA, Dey D, Leipsic J, Botker HE, Jensen JM. et al. Coronary plaque quantification and fractional flow reserve by coronary computed tomography angiography identify ischaemia-causing lesions. Eur Heart J 2016;37:1220–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. Shmilovich H, Cheng VY, Tamarappoo BK, Dey D, Nakazato R, Gransar H. et al. Vulnerable plaque features on coronary CT angiography as markers of inducible regional myocardial hypoperfusion from severe coronary artery stenoses. Atherosclerosis 2011;219:588–95. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. Diaz-Zamudio M, Dey D, Schuhbaeck A, Nakazato R, Gransar H, Slomka PJ. et al. Automated quantitative plaque burden from coronary CT angiography noninvasively predicts hemodynamic significance by using fractional flow reserve in intermediate coronary lesions. Radiology 2015;276:408–15. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Narula J, Nakano M, Virmani R, Kolodgie FD, Petersen R, Newcomb 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]
- 26. Hell MM, Dey D, Marwan M, Achenbach S, Schmid J, Schuhbaeck A.. Non- invasive prediction of hemodynamically significant coronary artery stenoses by contrast density difference in coronary CT angiography. Eur J Radiol 2015;84: 1502–8. [DOI] [PubMed] [Google Scholar]
- 27. Lansky AJ, Ng VG, Maehara A, Weisz G, Lerman A, Mintz GS. et al. Gender and the extent of coronary atherosclerosis, plaque composition, and clinical outcomes in acute coronary syndromes. JACC Cardiovasc Imaging 2012;5(3 Suppl):S62–72. [DOI] [PubMed] [Google Scholar]