- Email: [email protected]
SYNTAX Score Derived From Coronary CT Angiography for Prediction of Complex Percutaneous Coronary Interventions
ARTICLE IN PRESS
Original Investigation
SYNTAX Score Derived From Coronary CT Angiography for Prediction of Complex Percutaneous Coronary Interventions Aryeh Shalev, MD, Ryo Nakazato, MD, PhD, FACC, Reza Arsanjani, MD, Rine Nakanishi, MD, PhD, Hyung-Bok Park, MD, Yuka Otaki, MD, PhD, Victor Y. Cheng, MD, Heidi Gransar, MS, Troy M. LaBounty, MD, Sean W. Hayes, MD, MB,ChB, FRACP, Daniel S. Berman, MD, FACC, James K. Min, MD, FACC Abbreviations and Acronyms CABG Coronary artery bypass graft CAD Coronary artery disease CCTA Coronary computed tomographic angiography CIN Contrast-induced nephropathy ICA Invasive coronary angiography PCI Percutaneous coronary intervention
Rationale and Objectives: SYNTAX score is a useful metric determined at the time of invasive coronary angiography (ICA) to assess the complexity of coronary artery disease, and improves prediction of complications at the time of percutaneous complex intervention (PCI). We aimed to determine whether SYNTAX score can be reliably determined from coronary computed tomography angiography (CCTA) and whether a CCTA-derived SYNTAX score can predict complex PCI. Materials and Methods: SYNTAX scores were calculated on per-patient, per-vessel, and persegment basis in 154 consecutive patients who underwent CCTA and ICA. PCI complexity in 113 patients who underwent intervention was defined by total fluoroscopy time and contrast volume. Results: Compared to ICA, CCTA detected 285 of 302 (94%) obstructive lesions in 230 vessels, for which PCI was performed for 154 lesions in 131 vessels. Overall, on a per-patient basis, ICA-derived SYNTAX score was lower in comparison to CCTA-derived score (10.2 ± 8.0 vs 10.9 ± 8.3, P = 0.001). As compared to lesions in the lowest CCTA-derived segmental SYNTAX tertile, lesions in the highest tertile required longer fluoroscopy time (17.5 ± 12 min vs 11.5 ± 7.9 min, P = 0.01) and greater contrast volume (215.4 ± 125.5 mL vs 144.3 ± 49 mL, P = 0.02). Conclusion: SYNTAX scores derived from CCTA are concordant with those derived from ICA and correspond with complex PCI. Key Words: Multidetector computed tomography; coronary angiography; percutaneous coronary intervention; coronary artery disease; atherosclerotic plaque. © 2016 The Association of University Radiologists. Published by Elsevier Inc. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
INTRODUCTION Acad Radiol 2016; ■:■■–■■ From the Departments of Imaging (Division of Nuclear Medicine) and Medicine (Division of Cardiology), Cedars-Sinai Medical Center, Cedars-Sinai Heart Institute, Los Angeles, California (A.S., R.N., R.A., R.N., H.-B.P., Y.O., H.G., S.W.H., D.S.B.); Oklahoma Heart Institute, Tulsa, Oklahoma (V.Y.C.); Departments of Medicine and Radiology, University of Michigan, Ann Harbor, Michigan (T.M.LB.); Department of Medicine, David Geffen UCLA School of Medicine, University of California Los Angeles, Los Angeles, California (D.S.B.); Dalio Institute of Cardiovascular Imaging, NewYork-Presbyterian Hospital, 413 E 69th Street, BRB 108, New York, NY 10021 (J.K.M.). Received March 7, 2016; revised July 7, 2016; accepted July 15, 2016. Funding: This study has received funding by the National Heart, Lung, and Blood Institute of the National Institutes of Health (Bethesda, MD) under Award numbers R01HL115150 and R01HL118019 and also in part by a generous gift from the Dalio Institute of Cardiovascular Imaging (New York, New York) and the Michael Wolk Heart Foundation. Address correspondence to: J.K.M. e-mail: [email protected] © 2016 The Association of University Radiologists. Published by Elsevier Inc. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). http://dx.doi.org/10.1016/j.acra.2016.07.003
The SYNTAX score is a comprehensive angiographic scoring system commonly used to assess the complexity of coronary artery disease (CAD) by invasive coronary angiography (ICA). The SYNTAX score combines several validated angiographic classifications of CAD complexity based on the location and morphology of obstructive coronary lesions, and allows for robust risk stratification as well as guides appropriate methods of coronary revascularization (1). Coronary computed tomographic angiography (CCTA) has emerged as a non-invasive test that demonstrates high diagnostic accuracy compared to ICA for identification and exclusion of obstructive coronary stenosis, as well as plaque location, distribution, and extent (2–5). To date, whether a CCTA-derived SYNTAX score compares favorably to an ICA-derived SYNTAX score has not been 1
SHALEV ET AL
well evaluated. Further, whether these non-invasive CCTA findings can be applied in the pre-procedural setting to predict the complexity of percutaneous coronary intervention (PCI) is unknown. We sought to determine the feasibility and accuracy of CCTA-derived SYNTAX score when compared to an ICAderived SYNTAX score as a reference standard. Further, we determined the relationship of a CCTA-derived segmental SYNTAX score to the complexity of PCI. MATERIALS AND METHODS Study Population
We retrospectively studied consecutive patients who met the following inclusion criteria: (1) underwent CCTA and ICA within 30 days without intervening clinical cardiac event in the interscan period between CCTA and ICA; (2) were identified by CCTA as having an obstructive coronary stenosis >50% in non-stented segments; (3) no prior coronary artery bypass surgery (CABG). Among 170 eligible patients, 16 were excluded due to severe motion artifacts on CCTA. Before CCTA, we prospectively collected information on the presence of CAD risk factors. Hypertension was defined as a history of high blood pressure or treatment with antihypertensive medications. Diabetes mellitus was defined by previously made diagnosis and use of insulin or hypoglycemic agents. Dyslipidemia was defined as known but untreated dyslipidemia or current treatment with lipid-lowering medications. A smoking history was defined as current smoking or cessation within 3 months of testing. The study protocol was approved by the Cedars-Sinai Institutional Review Board. All participating patients provided informed consent for the use of their clinical and imaging data. CCTA Image Acquisition
CCTA scans were performed on a dual-source CT scanner (Somatom Definition, Siemens Medical Systems, Forchheim, Germany). The CCTA acquisition protocol is detailed in the Appendix. CCTA Image Reconstruction and Evaluation
Reconstruction of CCTA data was routinely performed at mid-diastole and, for CCTAs acquired by retrospective electrocardiogram helical methods, at end-systole if needed. Two experienced level III equivalent readers, blinded to patient characteristics and ICA findings, visually inspected and assessed all coronary segments >1.5 mm in diameter for presence of an obstructive stenosis, defined by a ≥50% luminal diameter narrowing (6–8). Each obstructive lesion ≥50% was further assessed and scored in accordance with the previously described SYNTAX score (1). CCTA image reconstruction and evaluation protocols, as well as calculation of CCTAderived SYNTAX scores are detailed in the Appendix. Disagreement between the two readers was resolved by means 2
Academic Radiology, Vol ■, No ■■, ■■ 2016
of consensus. A representative example of CCTA-based SYNTAX scoring can be seen in Figure 1. ICA Image Acquisition and Evaluation
ICA by standard catheterization techniques was performed (Innova, GE Healthcare, Buckinghamshire, United Kingdom). ICA images were analyzed by dedicated commercially available software (AGFA Heartlab, Greenville, SC). Two independent expert cardiologists performed image evaluation blinded to CCTA results, clinical information, and whether a PCI was performed. Each coronary segment >1.5 mm in diameter was visually inspected for presence of an obstructive lesion in a similar manner to CCTA interpretation. Each obstructive lesion ≥50% was further assessed and scored in accordance with the previously described SYNTAX score (1). Disagreement between the two readers was resolved by means of consensus, with independent adjudication of all obstructive lesions by CCTA and ICA to ensure crosscorrespondence and valid comparison of lesions between the modalities. Complexity of PCI
There were 14 interventional cardiologists with at least 10 years of experience (range 10–25 years) who performed the PCIs in the studied population. All interventionalists were blinded appropriately to CCTA SYNTAX scores. Complexity of PCI was determined by two metrics: (1) assessment of total fluoroscopy time of ICA and PCI; (2) volume of contrast use, inclusive of both the diagnostic and interventional portions of the procedure. Importantly, total fluoroscopy time was employed as a surrogate for overall radiation dose as no reliable metric exists for calculation of effective radiation dose from clinically performed ICA or PCI. Similarly, volume of contrast use was employed as a surrogate marker for risk of contrast-induced nephropathy (CIN). We performed analyses for individuals undergoing serial serum creatinine measurements. Creatinine levels before and within 96 hours after PCI were employed for evaluation. Baseline glomerular filtration rate was calculated by the Modification of Diet in Renal Disease formula (9). CIN was classified as grades 0, 1, and 2, as previously described in the literature (10). For metric (1) and (2), we assessed the impact of SYNTAX score among all patients who underwent PCI, as well as reserved analysis to cases that involved only a single lesion percutaneous revascularization (n = 85 patients). This was done to adequately estimate the impact of a segmental SYNTAX score on fluoroscopy time and contrast use. In exploratory analyses, procedure failure was also evaluated and defined as an inability to successfully cross an obstructive lesion with an intracoronary wire or post-procedural thrombolysis in myocardial infarction flow <3. Procedure-related complications included any serious adverse event, including occurrence of coronary dissection or perforation.
Academic Radiology, Vol ■, No ■■, ■■ 2016
CCTA-DERIVED SYNTAX SCORE AND COMPLEX PCI
Figure 1. SYNTAX score calculation by ICA and CCTA. ICA image is depicted in panel (a) and CCTA images are depicted in panels (b) and (c). The proximal left anterior descending and distal circumflex arteries are totally occluded by predominantly non-calcified plaques. Additionally, a mixed non-calcified or calcified plaque is causing >50% stenosis in the proximal obtuse marginal artery. SYNTAX score by CCTA is 27, whereas SYNTAX score by ICA is 28. CCTA, coronary computed tomographic angiography; ICA, invasive coronary angiography.
Statistical Methods
CCTA- and ICA-derived SYNTAX scores were compared for lesions considered anatomically obstructive by both modalities. Continuous variables are presented as means ± SD or medians and interquartile ranges for skewed variable values. Categorical variables were assessed using Pearson chi-square test or Fisher exact test for cell counts less than 6, and for ordered comparisons across >2 groups using the chi-square test for trend. Continuous variables were compared using 2-sample t test or one-way analysis of variance test across >2 groups. Cuzick test for trend was used for >2 ordered groups. Inter-observer and intra-observer variability studies were performed on CCTAs or ICAs, and assessed by kappa tests. Strength of agreement with kappa values was defined as follows: <0 none, 0–0.20 slight, 0.21–0.40 fair, 0.41–0.60 moderate, 0.61–0.80 good, 0.81–1.00 excellent. Agreement and correlations between ICA and CCTA were determined with Bland-Altman plots and Pearson correlation coefficients, respectively. Logistic regression was employed to assess the ability of SYNTAX tertiles to predict complex PCI. In addition, we performed a post hoc power analysis for the comparison of
SYNTAX scores on a per-patient basis. The post hoc analysis to detect a difference for a dichotomous classification of complex vs non-complex PCI on a per-patient basis yielded a power of 0.8 for CCTA. Two-tailed P < 0.05 was considered statistically significant. All data were analyzed using Stata version 11 (Stata Corp., College Station, TX).
RESULTS Patient Characteristics
There were 154 patients (66% males, mean age 66 ± 12 years) who had CCTA and ICA within 30 days, with baseline characteristics summarized in Table 1. In total, 6% of patients did not have obstructive CAD, as evidenced by ICA, whereas 45%, 38%, and 11% had 1-, 2-, and 3-vessel CAD, respectively. CAD distribution included left main (8%), left anterior descending (70%), left circumflex (34%), and right (37%) coronary arteries. Calcified plaques accounted for 13% of obstructive lesions, whereas mixed-calcified and non-calcified plaques were encountered in 57% and 30% of lesions, respectively. 3
Academic Radiology, Vol ■, No ■■, ■■ 2016
SHALEV ET AL
TABLE 3. SYNTAX CAD Variables: Comparison Between ICA and CCTA for 285 Lesions Considered Obstructive by Both
TABLE 1. Baseline Characteristics n = 154 Age (years)* Male gender Hypertension Dyslipidemia Diabetes mellitus Current cigarette smoking Family history of premature CAD BMI (Kg/m2)* Previous MI Previous PCI Baseline creatinine (mg/dL)* Baseline glomerular filtration rate* (mL/min/1.73 m2) Agatston coronary calcium score (n = 127)†
66 ± 12 107 (69) 101 (65) 118 (76) 38 (24) 30 (19) 62 (40) 27 ± 5 16 (10) 22 (14) 1.07 ± 0.9 85 ± 27
SYNTAX Components*
377 (148–940)
BMI, body mass index; CAD, coronary artery disease; MI, myocardial infarction; PCI, percutaneous coronary intervention. * mean ± SD, † median (interquartile range).
Agreement and Reproducibility of CCTA- and ICADerived SYNTAX Scores
There were 302 obstructive lesions detected by ICA and CCTA, with excellent agreement between CCTA and ICA for the number of obstructive lesions (kappa = 0.91; 95% confidence interval 0.88–0.93). Inter- and intra-observer variability for each modality showed significant agreement for number of obstructive lesions, bifurcation lesions, total occlusions, and SYNTAX tertiles (Table 2). The average length of totally occluded segments by CCTA was 17.7 ± 5.2 mm vs 15.8 ± 4.9 mm by ICA (P = 0.14). Overall, SYNTAX score elements for CCTA and ICA were compared for 285 lesions in 230 vessels considered obstructive by both (Table 3). Of note, the prevalence of intracoronary thrombi by ICA was extremely low in our patient population (two cases) and was undetected by CCTA. Incorporating intracoronary thrombi in the SYNTAX score results in minimal change (1 point) of the total SYNTAX score and did not affect the average SYNTAX score. The majority of lesions were estimated to cause ≥70% luminal narrowing (76% of lesions by CCTA vs 70% by ICA, P = 0.09; kappa = 0.62, 95%
Identified by ICA
Correctly Diagnosed by CCTA (%)
Agreement Between ICA and CCTA
36 19
34 (94) 9 (47)
0.94 (0.90–0.98) 0.63 (0.42–0.84)
10
9 (90)
0.65 (0.44–0.86)
14 14
0 (0) 9 (64)
0 0.68 (0.47–0.89)
7
5 (57)
0.66 (0.38–0.94)
73
53 (76)
0.56 (0.46–0.66)
53 28
31 (58) 24 (86)
0.53 (0.38–0.69) 0.76 (0.63–0.89)
2
0 (0)
0
19
13 (68)
0.63 (0.42–0.83)
Total occlusion Bridging collaterals Compromised side branch Blunt stump Segment visualized beyond occlusion Aorto-ostial lesion Bifurcation/ trifurcation Length >20 mm Heavy calcification Intracoronary thrombus Diffuse disease
CAD, coronary artery disease; CCTA, coronary computed tomographic angiography; ICA, invasive coronary angiography. Agreement is presented as kappa values (95% confidence intervals). * The following variables were not included for comparison given their very low prevalence: severe vessel tortuosity and side branch angulation (n = 4 cases for each).
confidence interval 0.52–0.72). Mean per-patient SYNTAX score derived from ICA was lower in comparison to mean CCTA-derived per-patient SYNTAX score (10.2 ± 8.0 vs 10.9 ± 8.3, P = 0.001). The SYNTAX tertiles for CCTA and ICA, on a per-patient basis, were coincidentally identical and defined as low-SYNTAX ≤ 7, 7 < intermediate-SYNTAX ≤ 12, and high-SYNTAX > 12. Correlation and agreement between CCTA- and ICA-derived SYNTAX
TABLE 2. Reproducibility Analyses—Inter-Observer Variability Between ICA Readers, CCTA Readers, as Well as Intra-observer Variability for Each of the CCTA Readers SYNTAX Score Tertiles* Inter-observer variability ICA CCTA Intra-observer variability CCTA 1st reader CCTA 2nd reader
Number of Obstructive Lesions
Bifurcation Lesions
Total Occlusions
0.81 (0.71–0.92) 0.85 (0.76–0.94)
0.94 (0.91–0.97) 0.95 (0.93–0.96)
0.64 (0.55–0.74) 0.71 (0.59–0.84)
0.96 (0.91–1.0) 0.95 (0.9–1.0)
0.85 (0.69–1.0) 0.78 (0.61–0.94)
0.94 (0.90–0.99) 0.98 (0.96–1.0)
0.71 (0.53–0.90) 0.79 (0.63–0.95)
1.0 0.86 (0.67–1.0)
CCTA, coronary computed tomographic angiography; ICA, invasive coronary angiography. * Data presented as weighted kappa values (95% confidence intervals).
4
Academic Radiology, Vol ■, No ■■, ■■ 2016
CCTA-DERIVED SYNTAX SCORE AND COMPLEX PCI
Figure 2. Correlation between CCTA- and ICA-derived SYNTAX scores on a per-segment, per-vessel, and per-patient basis. CCTA, coronary computed tomographic angiography; ICA, invasive coronary angiography.
scores, on per-patient, per-vessel, and per-segment basis, are presented in Figures 2 and 3a, respectively. Comparison Between Patients Referred to PCI vs CABG or Medical Therapy
The therapeutic decision was taken ad hoc by the operator and was based upon visual assessment of stenosis severity or functional assessment by prior nuclear myocardial perfusion imaging or functional flow reserve at the time of diagnostic angiography or intravascular ultrasound assessment. CCTA, ICA, and PCI were performed in 113 patients for 154 lesions in 131 vessels within less than 30 days between CCTA and PCI. Diagnostic ICA and PCI were performed on the same day for the majority of the patients (95%). Visual assessment of stenosis severity was associated with coronary intervention in 130 (84%) obstructive lesions, whereas functional assessment accounted for the remainder of interventions. There were 15 patients referred for CABG for 48 lesions. Diffuse obstructive disease was associated with the decision
to operate in 13 patients, whereas 2 patients underwent concomitant valve replacement and CABG. Within each therapeutic approach group, the per-patient CCTA- and ICA-derived SYNTAX scores, as well as segmental SYNTAX scores, were similar (Table 4). As expected, the per-patient SYNTAX scores in patients who were referred to surgery were significantly higher in comparison to patients that were referred to PCI or medical management. The segmental SYNTAX scores of CABG lesions were higher in comparison to lesions that were designated to medical management but were similar in comparison PCI lesions (Table 4). Secondary Analysis of Patients Undergoing PCI
CCTA- and ICA-derived per-patient and per-vessel SYNTAX scores demonstrated excellent agreement (Fig 3b) and correlation (Pearson correlation coefficients were 0.99, 0.95, 0.9, and 0.93 for left main (LM), left anterior descending (LAD), left circumflex (LCX), and right coronary artery (RCA), respectively; P < 0.05 for all), with CCTA correctly detecting 5
SHALEV ET AL
Academic Radiology, Vol ■, No ■■, ■■ 2016
Figure 3. Bland-Altman plots for agreement for the (a) entire study cohort and (b) patients undergoing PCI.
6
Academic Radiology, Vol ■, No ■■, ■■ 2016
TABLE 4. CCTA- and ICA-derived Per-patient and Segmental SYNTAX Scores According to Therapeutic Approach Per Patient SYNTAX Score
Medical Management (n = 26 Patients)
PCI (n = 113)
CABG (n = 15)
CCTA ICA Segmental SYNTAX score CCTA ICA
6.8 ± 4.1 7.2 ± 5.4 Medical management (n = 78 lesions) 4.9 ± 3.7 5.0 ± 3.8
10.3 ± 6.3 10.1 ± 6.1 PCI (n = 154 lesions) 5.3 ± 3.5 5.5 ± 3.5
23.5 ± 14.6*,† 21.7 ± 12.1*,† CABG (n = 48 lesions) 6.4 ± 4.5* 6.5 ± 4.4*
CCTA, coronary computed tomographic angiography; CABG, coronary artery bypass graft; ICA, invasive coronary angiography; PCI, percutaneous coronary intervention. * P value <0.05 for comparison of CABG vs medical management. † P value <0.05 for comparison of CABG vs PCI.
all the lesions that led to revascularization. On a per patient basis, SYNTAX scores derived from CCTA vs ICA did not differ (10.3 ± 6.3 vs 10.1 ± 6.1, P = 0.3). For patients undergoing PCI, CCTA-, and ICA-derived segmental SYNTAX scores were similar (5.5 ± 3.6 vs 5.5 ± 3.5, P = 0.7). Similarly, for patients with single-lesion PCI, mean CCTA-, and ICAderived segmental SYNTAX scores were similar (5.9 ± 3.6 vs 5.9 ± 3.5, respectively). Total Contrast Volume
Total contrast volume was moderately correlated with CCTAand ICA-derived segmental SYNTAX score tertiles (r = 0.50, P < 0.05 and r = 0.51, P < 0.05, respectively). Distribution of contrast volumes by SYNTAX tertiles is presented in Table 5.
CCTA-DERIVED SYNTAX SCORE AND COMPLEX PCI
The contrast volume administered at each of the per-patient or segmental SYNTAX score tertiles was not significantly different between CCTA- or ICA-derived SYNTAX tertiles. Ten patients had baseline glomerular filtration rate >0 and <60 mL/min and none developed CIN. Overall CIN grades 1 or 2 were observed in three patients and one patient, respectively. CCTA- and ICA-derived segmental SYNTAX score as a function of contrast media volume was similar among patients who developed CIN as compared to patients who did not (270 ± 95 mL vs 190 ± 93 mL, P = 0.2). Total Fluoroscopy Time
Total fluoroscopy time was moderately correlated with CCTAand ICA-derived segmental SYNTAX score tertiles (r = 0.42, P < 0.05 and r = 0.43, P < 0.05, respectively). Distribution of total fluoroscopy time is presented in Table 5. Of note, there was no correlation between operators’ expertise and metrics of contrast volume utilization or fluoroscopy time. Procedure Failure or Complication
PCI was unsuccessful in four totally occluded lesions (3%). The mean CCTA-derived segmental SYNTAX score for lesions with unsuccessful interventions was significantly higher than for lesions with successful interventions (13 ± 8 vs 5.3 ± 3.2, P < 0.001), a finding in accord with ICA-derived segmental SYNTAX scores. Dissection occurred in additional three lesions (2%) with mean CCTA- and ICA-derived segmental SYNTAX scores of 3.0 ± 1.0 and 3.3 ± 1.1, respectively (P = 0.7). DISCUSSION In the present study we assessed the feasibility and accuracy of a CCTA-derived SYNTAX score to predict complex PCI.
TABLE 5. Total Fluoroscopy Time and Contrast Volume According to Total and Segmental SYNTAX Score Tertiles Among PCI Subgroup (n = 113)
Overall SYNTAX score ICA Contrast volume (mL) Total fluoroscopy time (min) CCTA Contrast volume (ml) Total fluoroscopy time (min) Segmental SYNTAX scores ICA Contrast volume (mL) Total fluoroscopy time (min) CCTA Contrast volume (mL) Total fluoroscopy time (min)
1st Tertile
2nd Tertile
3rd Tertile
P Value
153.4 ± 62.7 12.7 ± 9.2
193.0 ± 64.8 16.4 ± 9.8
251.7 ± 139.4 22.1 ± 14.6
<0.001 <0.001
152.7 ± 62.7 12.6 ± 9.0
188.6 ± 63.7 17.0 ± 10.4
251.8 ± 133.6 21.5 ± 14.2
<0.001 <0.001
135.7 ± 43.2 10.6 ± 4.8
184.2 ± 59.2 17 ± 11.6
222.7 ± 127.1 18.7 ± 12.5
<0.01 0.02
144.3 ± 49 11.5 ± 7.9
194.3 ± 68.4 19.2 ± 11.6
215.4 ± 125.5 17.5 ± 12
0.01 0.03
Mean ± SD
Mean ± SD
CCTA, coronary computed tomographic angiography; ICA, invasive coronary angiography; PCI, percutaneous coronary intervention.
7
SHALEV ET AL
Within the study population, we noted generally high concordance and diagnostic accuracy of the CCTA-derived SYNTAX score as compared to an ICA-derived SYNTAX score reference standard. Further, higher CCTA-derived SYNTAX scores were associated with increasing complexity of coronary revascularization, with higher volumes of iodinated contrast, longer fluoroscopy times, and greater use of advanced devices and procedures. The SYNTAX score is a useful ICA score that incorporates anatomical and morphological lesion characteristics that can be used to reliably determine the complexity of CAD (1,8). SYNTAX score is well-validated, highly reproducible (11,12), and predictive of adverse cardiovascular events in a myriad of scenarios concerning PCI in multivessel and left main CAD (13–15). The SYNTAX score provides greater discriminatory ability compared to the modified American College of Cardiology/American Heart Association lesion classification system (13). CCTA has been established as a non-invasive anatomic test that offers high diagnostic performance for detection and exclusion of high-grade coronary stenosis when compared to traditional ICA. Moreover, CCTA allows for direct visualization of additional CAD features beyond stenosis severity, including measures of lesion length, lesion location (eg, at bifurcations of coronary vessels), and a multitude of other variables commonly used in the ICA SYNTAX (16). Several of these findings of CAD extent, severity, location, and morphology manifest important prognostic implications, and are associated with future death and major adverse cardiovascular events across numerous patient populations (17–22). As previously demonstrated by our investigative group, proximal LAD stenosis, coronary plaque distribution, weighted segmental stenosis scores, and number of vessels with ≥50% and ≥70% stenosis strongly predict mortality (17). Papadopoulou et al. have studied the application of CCTA-derived SYNTAX score in a group of 80 patients, and concluded that CCTA-derived score is feasible, reproducible, and comparable to ICA-derived SYNTAX score (23). Our present study is both in accordance with as well as advances these findings and, to our knowledge, represents the largest study to date to assess the reliability and applicability of a CCTA-derived SYNTAX score. CCTA- and ICA-derived SYNTAX scores demonstrated generally high concordance on a per-segment, per-vessel, and perpatient basis, as well as substantial agreement for the majority of the SYNTAX score variables. CCTA and ICA were in excellent agreement for detection of obstructive lesions and total occlusions, but only in moderate agreement for lesion length. The potential reasons for this are manifold. Given the ability of CCTA to define atherosclerosis in three dimensions rather than two, it is possible that multiplanar reconstructions of coronary vessels by CCTA may provide slightly different measures over ICA which may be susceptible to foreshortening and vessel overlap. Although determination of heavy calcification by CCTA was in good agreement with ICA, severe coronary calcification was prominently visualized and noted as a limitation of currentgeneration CCTA for measures of stenosis severity; as such, this 8
Academic Radiology, Vol ■, No ■■, ■■ 2016
may have led to overestimation by CCTA. Blunt stumps and intracoronary thrombi were completely undetected by CCTA. These disparities may stem from the superior spatial and temporal resolution of ICA over CCTA, and may suggest a need for improved methods of CCTA interpretation. Finally, additional elements of the SYNTAX score, ie, tortuous vessels, were rarely encountered in the study population, were therefore not compared between CCTA and ICA, and underscore the need for future larger studies. Perhaps more important than the demonstration that CCTAderived SYNTAX scores are accurate and concordant with ICA-derived scores is the finding that this non-invasive method is associated with PCI complexity. Although prior studies have reported the usefulness of CCTA to identify chronic total occlusions that are amenable to revascularization (24–27), these investigations are limited in number and scope, have not evaluated the totality of SYNTAX CAD variables, and have not systematically examined the utility of CCTA to identify complex PCI across a range of CAD states, as was done in this study. Stähli et al. have reported their experience with applying a CCTA-derived SYNTAX score for assessment of 66 lesions deemed eligible for intervention (28). The CCTA- and ICAderived SYNTAX scores were comparable and predictive of failed PCI, but not for PCI complexity. Our study directly extends these important observations by providing a comprehensive assessment of PCI complexity by an array of metrics that indicate heightened complexity of PCI. These factors may allow for pre-procedural assessment and planning in a more refined manner, and may allow for enhanced physicianpatient interactions for maximizing informed consent. As complex CAD is often accompanied by baseline chronic kidney disease, concerns regarding a heightened incidence of CIN exist. Further, radiation dose is an important consideration even within an aging population, and the stochastic risk of radiation-induced cancer remains of non-negligible concern. Complex procedures entail higher contrast volume and radiation exposure, as has been reported in an array of prior studies (29) and corroborated by the present study findings. In the present study, CCTA- and ICA-derived segmental SYNTAX scores correlated well between patients with and without CIN. This finding may be largely explained by the relatively low CIN incidence and thus, whether SYNTAX score can predict development of CIN in larger cohorts remains to be determined. This study is not without limitations. While all patients undergoing CCTA were prospectively identified and enrolled, examination of the ICA and PCI of these generally lowintermediate risk patients was performed in a retrospective fashion, and potentially affected by selection or referral bias. Moreover, the study population represented low-intermediate risk individuals who underwent ICA based upon CCTA and thus, whether these findings are applicable to higher risk patients requires further study. As CCTA is prone to overestimation of stenosis severity, the overall accuracy of the CCTA-derived scores may be reduced (3–5,30,31). Further, prior fractional flow reserve
Academic Radiology, Vol ■, No ■■, ■■ 2016
studies have proven the prognostic utility of a “functional” SYNTAX score, and this was not addressed in the current investigation (32). As radiation dose by ICA is difficult to quantify, we relied upon estimates based upon total fluoroscopy time as a surrogate for total effective radiation dose. Finally, as PCIs are primarily performed in an ad hoc manner by interventionalists of potentially different skill levels for patients with various comorbidities, CCTA-based SYNTAX scores cannot alone serve as the definitive guide for prediction of individuals who may experience complex coronary interventions. In conclusion, SYNTAX scores derived from CCTA are concordant with those derived from ICA, and correspond well with complex PCI.
REFERENCES 1. Sianos G, Morel M-A, Kappetein AP, et al. The SYNTAX Score: an angiographic tool grading the complexity of coronary artery disease. EuroIntervention 2005; 1:219–227. 2. Raff GL, Gallagher MJ, Oein O, et al. Diagnostic accuracy of noninvasive coronary angiography using 64-slice spiral computed tomography. J Am Coll Cardiol 2005; 46:552–557. 3. Miller JM, Rochitte CE, Dewey M, et al. Diagnostic performance of coronary angiography by 64-row CT. N Engl J Med 2008; 359:2324–2336. 4. Budoff MJ, Dowe D, Jollis JG, et al. Diagnostic performance of 64multidetector row coronary computed tomographic angiography for evaluation of coronary artery stenosis in individuals without known coronary artery disease: results from the prospective multicenter ACCURACY (Assessment by Coronary Computed Tomographic Angiography of Individuals Undergoing Invasive Coronary Angiography) trial. J Am Coll Cardiol 2008; 52:1724–1732. 5. Meijboom WB, Meijs MFL, Schuijf JD, et al. Diagnostic accuracy of 64slice computed tomography coronary angiography: a prospective, multicenter, multivendor study. J Am Coll Cardiol 2008; 52:2135–2144. 6. Ferencik M, Ropers D, Abbara S, et al. Diagnostic accuracy of image postprocessing methods for the detection of coronary artery stenoses by using multidetector CT. Radiology 2007; 243:696–702. 7. Cheng V, Gutstein A, Wolak A, et al. Moving beyond binary grading of coronary arterial stenoses on coronary computed tomographic angiography: insights for the imager and referring clinician. JACC Cardiovasc Imaging 2008; 1:460–471. 8. Leaman DM, Brower RW, Meester GT, et al. Coronary artery atherosclerosis: severity of the disease, severity of angina pectoris and compromised left ventricular function. Circulation 1981; 63:285–299. 9. Levey AS, Stevens LA, Schmid CH, et al. A new equation to estimate glomerular filtration rate. Ann Intern Med 2009; 150:604–612. 10. Harjai KJ, Raizada A, Shenoy C, et al. A comparison of contemporary definitions of contrast nephropathy in patients undergoing percutaneous coronary intervention and a proposal for a novel nephropathy grading system. Am J Cardiol 2008; 101:812–819. 11. Serruys PW, Onuma Y, Garg S, et al. Assessment of the SYNTAX score in the Syntax study. EuroIntervention 2009; 5:50–56. 12. Garg S, Girasis C, Sarno G, et al. The SYNTAX score revisited: a reassessment of the SYNTAX score reproducibility. Catheter Cardiovasc Interv 2010; 75:946–952. 13. Valgimigli M, Serruys PW, Tsuchida K, et al. Cyphering the complexity of coronary artery disease using the syntax score to predict clinical outcome in patients with three-vessel lumen obstruction undergoing percutaneous coronary intervention. Am J Cardiol 2007; 99:1072–1081. 14. Serruys PW, Morice M-C, Kappetein AP, et al. Percutaneous coronary intervention versus coronary-artery bypass grafting for severe coronary artery disease. N Engl J Med 2009; 360:961–972. 15. Wykrzykowska JJ, Garg S, Girasis C, et al. Value of the SYNTAX score for risk assessment in the all-comers population of the randomized multicenter LEADERS (Limus Eluted from a durable versus ERodable stent coating) trial. J Am Coll Cardiol 2010; 56:272–277.
CCTA-DERIVED SYNTAX SCORE AND COMPLEX PCI
16. Voros S, Rinehart S, Qian Z, et al. Coronary atherosclerosis imaging by coronary CT angiography: current status, correlation with intravascular interrogation and meta-analysis. JACC Cardiovasc Imaging 2011; 4:537– 548. 17. Min JK, Shaw LJ, Devereux RB, et al. Prognostic value of multidetector coronary computed tomographic angiography for prediction of allcause mortality. J Am Coll Cardiol 2007; 50:1161–1170. 18. Hulten EA, Carbonaro S, Petrillo SP, et al. Prognostic value of cardiac computed tomography angiography: a systematic review and metaanalysis. J Am Coll Cardiol 2011; 57:1237–1247. 19. Min JK, Dunning A, Lin FY, et al. Age- and sex-related differences in allcause mortality risk based on coronary computed tomography angiography findings results from the International Multicenter CONFIRM (COroNary CT Angiography Evaluation For Clinical Outcomes: An InteRnational Multicenter Registry) of 23,854 patients without known coronary artery disease. J Am Coll Cardiol 2011; 58:849–860. 20. Rana JS, Dunning A, Achenbach S, et al. Differences in prevalence, extent, severity, and prognosis of coronary artery disease among patients with and without diabetes undergoing coronary computed tomography angiography: results from 10,110 individuals from the CONFIRM (COroNary CT Angiography Evaluation For Clinical Outcomes: An InteRnational Multicenter Registry). Diabetes Care 2012; 35:1787–1794. 21. Chow BJW, Small G, Yam Y, et al. Incremental prognostic value of cardiac computed tomography in coronary artery disease using CONFIRM: COroNary CT Angiography Evaluation For Clinical Outcomes: An InteRnational Multicenter Registry. Circ Cardiovasc Imaging 2011; 4:463– 472. 22. Villines TC, Hulten EA, Shaw LJ, et al. Prevalence and severity of coronary artery disease and adverse events among symptomatic patients with coronary artery calcification scores of zero undergoing coronary computed tomography angiography: results from the CONFIRM (COroNary CT Angiography Evaluation For Clinical Outcomes: An InteRnational Multicenter Registry). J Am Coll Cardiol 2011; 58:2533–2540. 23. Papadopoulou S-L, Girasis C, Dharampal A, et al. CT-SYNTAX score: a feasibility and reproducibility study. JACC Cardiovasc Imaging 2013; 6:413– 415. 24. Mollet NR, Hoye A, Lemos PA, et al. Value of preprocedure multislice computed tomographic coronary angiography to predict the outcome of percutaneous recanalization of chronic total occlusions. Am J Cardiol 2005; 95:240–243. 25. Soon KH, Cox N, Wong A, et al. CT coronary angiography predicts the outcome of percutaneous coronary intervention of chronic total occlusion. J Interv Cardiol 2007; 20:359–366. 26. García-García HM, van Mieghem CAG, Gonzalo N, et al. Computed tomography in total coronary occlusions (CTTO registry): radiation exposure and predictors of successful percutaneous intervention. EuroIntervention 2009; 4:607–616. 27. Hsu JT, Kyo E, Chu CM, et al. Impact of calcification length ratio on the intervention for chronic total occlusions. Int J Cardiol 2011; 150:135– 141. 28. Stähli BE, Bonassin F, Goetti R, et al. Coronary computed tomography angiography indicates complexity of percutaneous coronary interventions. J Invasive Cardiol 2012; 24:196–201. 29. Wertman BM, Cheng VY, Kar S, et al. Characterization of complex coronary artery stenosis morphology by coronary computed tomographic angiography. JACC Cardiovasc Imaging 2009; 2:950–958. 30. Meijboom WB, Van Mieghem CAG, van Pelt N, et al. Comprehensive assessment of coronary artery stenoses: computed tomography coronary angiography versus conventional coronary angiography and correlation with fractional flow reserve in patients with stable angina. J Am Coll Cardiol 2008; 52:636–643. 31. Schuijf JD, Wijns W, Jukema JW, et al. Relationship between noninvasive coronary angiography with multi-slice computed tomography and myocardial perfusion imaging. J Am Coll Cardiol 2006; 48:2508–2514. 32. Nam C-W, Mangiacapra F, Entjes R, et al. Functional SYNTAX score for risk assessment in multivessel coronary artery disease. J Am Coll Cardiol 2011; 58:1211–1218.
SUPPLEMENTARY DATA Supplementary data related to this article can be found at http://dx.doi.org/10.1016/j.acra.2016.07.003.
9
Original Investigation
SYNTAX Score Derived From Coronary CT Angiography for Prediction of Complex Percutaneous Coronary Interventions Aryeh Shalev, MD, Ryo Nakazato, MD, PhD, FACC, Reza Arsanjani, MD, Rine Nakanishi, MD, PhD, Hyung-Bok Park, MD, Yuka Otaki, MD, PhD, Victor Y. Cheng, MD, Heidi Gransar, MS, Troy M. LaBounty, MD, Sean W. Hayes, MD, MB,ChB, FRACP, Daniel S. Berman, MD, FACC, James K. Min, MD, FACC Abbreviations and Acronyms CABG Coronary artery bypass graft CAD Coronary artery disease CCTA Coronary computed tomographic angiography CIN Contrast-induced nephropathy ICA Invasive coronary angiography PCI Percutaneous coronary intervention
Rationale and Objectives: SYNTAX score is a useful metric determined at the time of invasive coronary angiography (ICA) to assess the complexity of coronary artery disease, and improves prediction of complications at the time of percutaneous complex intervention (PCI). We aimed to determine whether SYNTAX score can be reliably determined from coronary computed tomography angiography (CCTA) and whether a CCTA-derived SYNTAX score can predict complex PCI. Materials and Methods: SYNTAX scores were calculated on per-patient, per-vessel, and persegment basis in 154 consecutive patients who underwent CCTA and ICA. PCI complexity in 113 patients who underwent intervention was defined by total fluoroscopy time and contrast volume. Results: Compared to ICA, CCTA detected 285 of 302 (94%) obstructive lesions in 230 vessels, for which PCI was performed for 154 lesions in 131 vessels. Overall, on a per-patient basis, ICA-derived SYNTAX score was lower in comparison to CCTA-derived score (10.2 ± 8.0 vs 10.9 ± 8.3, P = 0.001). As compared to lesions in the lowest CCTA-derived segmental SYNTAX tertile, lesions in the highest tertile required longer fluoroscopy time (17.5 ± 12 min vs 11.5 ± 7.9 min, P = 0.01) and greater contrast volume (215.4 ± 125.5 mL vs 144.3 ± 49 mL, P = 0.02). Conclusion: SYNTAX scores derived from CCTA are concordant with those derived from ICA and correspond with complex PCI. Key Words: Multidetector computed tomography; coronary angiography; percutaneous coronary intervention; coronary artery disease; atherosclerotic plaque. © 2016 The Association of University Radiologists. Published by Elsevier Inc. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
INTRODUCTION Acad Radiol 2016; ■:■■–■■ From the Departments of Imaging (Division of Nuclear Medicine) and Medicine (Division of Cardiology), Cedars-Sinai Medical Center, Cedars-Sinai Heart Institute, Los Angeles, California (A.S., R.N., R.A., R.N., H.-B.P., Y.O., H.G., S.W.H., D.S.B.); Oklahoma Heart Institute, Tulsa, Oklahoma (V.Y.C.); Departments of Medicine and Radiology, University of Michigan, Ann Harbor, Michigan (T.M.LB.); Department of Medicine, David Geffen UCLA School of Medicine, University of California Los Angeles, Los Angeles, California (D.S.B.); Dalio Institute of Cardiovascular Imaging, NewYork-Presbyterian Hospital, 413 E 69th Street, BRB 108, New York, NY 10021 (J.K.M.). Received March 7, 2016; revised July 7, 2016; accepted July 15, 2016. Funding: This study has received funding by the National Heart, Lung, and Blood Institute of the National Institutes of Health (Bethesda, MD) under Award numbers R01HL115150 and R01HL118019 and also in part by a generous gift from the Dalio Institute of Cardiovascular Imaging (New York, New York) and the Michael Wolk Heart Foundation. Address correspondence to: J.K.M. e-mail: [email protected] © 2016 The Association of University Radiologists. Published by Elsevier Inc. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). http://dx.doi.org/10.1016/j.acra.2016.07.003
The SYNTAX score is a comprehensive angiographic scoring system commonly used to assess the complexity of coronary artery disease (CAD) by invasive coronary angiography (ICA). The SYNTAX score combines several validated angiographic classifications of CAD complexity based on the location and morphology of obstructive coronary lesions, and allows for robust risk stratification as well as guides appropriate methods of coronary revascularization (1). Coronary computed tomographic angiography (CCTA) has emerged as a non-invasive test that demonstrates high diagnostic accuracy compared to ICA for identification and exclusion of obstructive coronary stenosis, as well as plaque location, distribution, and extent (2–5). To date, whether a CCTA-derived SYNTAX score compares favorably to an ICA-derived SYNTAX score has not been 1
SHALEV ET AL
well evaluated. Further, whether these non-invasive CCTA findings can be applied in the pre-procedural setting to predict the complexity of percutaneous coronary intervention (PCI) is unknown. We sought to determine the feasibility and accuracy of CCTA-derived SYNTAX score when compared to an ICAderived SYNTAX score as a reference standard. Further, we determined the relationship of a CCTA-derived segmental SYNTAX score to the complexity of PCI. MATERIALS AND METHODS Study Population
We retrospectively studied consecutive patients who met the following inclusion criteria: (1) underwent CCTA and ICA within 30 days without intervening clinical cardiac event in the interscan period between CCTA and ICA; (2) were identified by CCTA as having an obstructive coronary stenosis >50% in non-stented segments; (3) no prior coronary artery bypass surgery (CABG). Among 170 eligible patients, 16 were excluded due to severe motion artifacts on CCTA. Before CCTA, we prospectively collected information on the presence of CAD risk factors. Hypertension was defined as a history of high blood pressure or treatment with antihypertensive medications. Diabetes mellitus was defined by previously made diagnosis and use of insulin or hypoglycemic agents. Dyslipidemia was defined as known but untreated dyslipidemia or current treatment with lipid-lowering medications. A smoking history was defined as current smoking or cessation within 3 months of testing. The study protocol was approved by the Cedars-Sinai Institutional Review Board. All participating patients provided informed consent for the use of their clinical and imaging data. CCTA Image Acquisition
CCTA scans were performed on a dual-source CT scanner (Somatom Definition, Siemens Medical Systems, Forchheim, Germany). The CCTA acquisition protocol is detailed in the Appendix. CCTA Image Reconstruction and Evaluation
Reconstruction of CCTA data was routinely performed at mid-diastole and, for CCTAs acquired by retrospective electrocardiogram helical methods, at end-systole if needed. Two experienced level III equivalent readers, blinded to patient characteristics and ICA findings, visually inspected and assessed all coronary segments >1.5 mm in diameter for presence of an obstructive stenosis, defined by a ≥50% luminal diameter narrowing (6–8). Each obstructive lesion ≥50% was further assessed and scored in accordance with the previously described SYNTAX score (1). CCTA image reconstruction and evaluation protocols, as well as calculation of CCTAderived SYNTAX scores are detailed in the Appendix. Disagreement between the two readers was resolved by means 2
Academic Radiology, Vol ■, No ■■, ■■ 2016
of consensus. A representative example of CCTA-based SYNTAX scoring can be seen in Figure 1. ICA Image Acquisition and Evaluation
ICA by standard catheterization techniques was performed (Innova, GE Healthcare, Buckinghamshire, United Kingdom). ICA images were analyzed by dedicated commercially available software (AGFA Heartlab, Greenville, SC). Two independent expert cardiologists performed image evaluation blinded to CCTA results, clinical information, and whether a PCI was performed. Each coronary segment >1.5 mm in diameter was visually inspected for presence of an obstructive lesion in a similar manner to CCTA interpretation. Each obstructive lesion ≥50% was further assessed and scored in accordance with the previously described SYNTAX score (1). Disagreement between the two readers was resolved by means of consensus, with independent adjudication of all obstructive lesions by CCTA and ICA to ensure crosscorrespondence and valid comparison of lesions between the modalities. Complexity of PCI
There were 14 interventional cardiologists with at least 10 years of experience (range 10–25 years) who performed the PCIs in the studied population. All interventionalists were blinded appropriately to CCTA SYNTAX scores. Complexity of PCI was determined by two metrics: (1) assessment of total fluoroscopy time of ICA and PCI; (2) volume of contrast use, inclusive of both the diagnostic and interventional portions of the procedure. Importantly, total fluoroscopy time was employed as a surrogate for overall radiation dose as no reliable metric exists for calculation of effective radiation dose from clinically performed ICA or PCI. Similarly, volume of contrast use was employed as a surrogate marker for risk of contrast-induced nephropathy (CIN). We performed analyses for individuals undergoing serial serum creatinine measurements. Creatinine levels before and within 96 hours after PCI were employed for evaluation. Baseline glomerular filtration rate was calculated by the Modification of Diet in Renal Disease formula (9). CIN was classified as grades 0, 1, and 2, as previously described in the literature (10). For metric (1) and (2), we assessed the impact of SYNTAX score among all patients who underwent PCI, as well as reserved analysis to cases that involved only a single lesion percutaneous revascularization (n = 85 patients). This was done to adequately estimate the impact of a segmental SYNTAX score on fluoroscopy time and contrast use. In exploratory analyses, procedure failure was also evaluated and defined as an inability to successfully cross an obstructive lesion with an intracoronary wire or post-procedural thrombolysis in myocardial infarction flow <3. Procedure-related complications included any serious adverse event, including occurrence of coronary dissection or perforation.
Academic Radiology, Vol ■, No ■■, ■■ 2016
CCTA-DERIVED SYNTAX SCORE AND COMPLEX PCI
Figure 1. SYNTAX score calculation by ICA and CCTA. ICA image is depicted in panel (a) and CCTA images are depicted in panels (b) and (c). The proximal left anterior descending and distal circumflex arteries are totally occluded by predominantly non-calcified plaques. Additionally, a mixed non-calcified or calcified plaque is causing >50% stenosis in the proximal obtuse marginal artery. SYNTAX score by CCTA is 27, whereas SYNTAX score by ICA is 28. CCTA, coronary computed tomographic angiography; ICA, invasive coronary angiography.
Statistical Methods
CCTA- and ICA-derived SYNTAX scores were compared for lesions considered anatomically obstructive by both modalities. Continuous variables are presented as means ± SD or medians and interquartile ranges for skewed variable values. Categorical variables were assessed using Pearson chi-square test or Fisher exact test for cell counts less than 6, and for ordered comparisons across >2 groups using the chi-square test for trend. Continuous variables were compared using 2-sample t test or one-way analysis of variance test across >2 groups. Cuzick test for trend was used for >2 ordered groups. Inter-observer and intra-observer variability studies were performed on CCTAs or ICAs, and assessed by kappa tests. Strength of agreement with kappa values was defined as follows: <0 none, 0–0.20 slight, 0.21–0.40 fair, 0.41–0.60 moderate, 0.61–0.80 good, 0.81–1.00 excellent. Agreement and correlations between ICA and CCTA were determined with Bland-Altman plots and Pearson correlation coefficients, respectively. Logistic regression was employed to assess the ability of SYNTAX tertiles to predict complex PCI. In addition, we performed a post hoc power analysis for the comparison of
SYNTAX scores on a per-patient basis. The post hoc analysis to detect a difference for a dichotomous classification of complex vs non-complex PCI on a per-patient basis yielded a power of 0.8 for CCTA. Two-tailed P < 0.05 was considered statistically significant. All data were analyzed using Stata version 11 (Stata Corp., College Station, TX).
RESULTS Patient Characteristics
There were 154 patients (66% males, mean age 66 ± 12 years) who had CCTA and ICA within 30 days, with baseline characteristics summarized in Table 1. In total, 6% of patients did not have obstructive CAD, as evidenced by ICA, whereas 45%, 38%, and 11% had 1-, 2-, and 3-vessel CAD, respectively. CAD distribution included left main (8%), left anterior descending (70%), left circumflex (34%), and right (37%) coronary arteries. Calcified plaques accounted for 13% of obstructive lesions, whereas mixed-calcified and non-calcified plaques were encountered in 57% and 30% of lesions, respectively. 3
Academic Radiology, Vol ■, No ■■, ■■ 2016
SHALEV ET AL
TABLE 3. SYNTAX CAD Variables: Comparison Between ICA and CCTA for 285 Lesions Considered Obstructive by Both
TABLE 1. Baseline Characteristics n = 154 Age (years)* Male gender Hypertension Dyslipidemia Diabetes mellitus Current cigarette smoking Family history of premature CAD BMI (Kg/m2)* Previous MI Previous PCI Baseline creatinine (mg/dL)* Baseline glomerular filtration rate* (mL/min/1.73 m2) Agatston coronary calcium score (n = 127)†
66 ± 12 107 (69) 101 (65) 118 (76) 38 (24) 30 (19) 62 (40) 27 ± 5 16 (10) 22 (14) 1.07 ± 0.9 85 ± 27
SYNTAX Components*
377 (148–940)
BMI, body mass index; CAD, coronary artery disease; MI, myocardial infarction; PCI, percutaneous coronary intervention. * mean ± SD, † median (interquartile range).
Agreement and Reproducibility of CCTA- and ICADerived SYNTAX Scores
There were 302 obstructive lesions detected by ICA and CCTA, with excellent agreement between CCTA and ICA for the number of obstructive lesions (kappa = 0.91; 95% confidence interval 0.88–0.93). Inter- and intra-observer variability for each modality showed significant agreement for number of obstructive lesions, bifurcation lesions, total occlusions, and SYNTAX tertiles (Table 2). The average length of totally occluded segments by CCTA was 17.7 ± 5.2 mm vs 15.8 ± 4.9 mm by ICA (P = 0.14). Overall, SYNTAX score elements for CCTA and ICA were compared for 285 lesions in 230 vessels considered obstructive by both (Table 3). Of note, the prevalence of intracoronary thrombi by ICA was extremely low in our patient population (two cases) and was undetected by CCTA. Incorporating intracoronary thrombi in the SYNTAX score results in minimal change (1 point) of the total SYNTAX score and did not affect the average SYNTAX score. The majority of lesions were estimated to cause ≥70% luminal narrowing (76% of lesions by CCTA vs 70% by ICA, P = 0.09; kappa = 0.62, 95%
Identified by ICA
Correctly Diagnosed by CCTA (%)
Agreement Between ICA and CCTA
36 19
34 (94) 9 (47)
0.94 (0.90–0.98) 0.63 (0.42–0.84)
10
9 (90)
0.65 (0.44–0.86)
14 14
0 (0) 9 (64)
0 0.68 (0.47–0.89)
7
5 (57)
0.66 (0.38–0.94)
73
53 (76)
0.56 (0.46–0.66)
53 28
31 (58) 24 (86)
0.53 (0.38–0.69) 0.76 (0.63–0.89)
2
0 (0)
0
19
13 (68)
0.63 (0.42–0.83)
Total occlusion Bridging collaterals Compromised side branch Blunt stump Segment visualized beyond occlusion Aorto-ostial lesion Bifurcation/ trifurcation Length >20 mm Heavy calcification Intracoronary thrombus Diffuse disease
CAD, coronary artery disease; CCTA, coronary computed tomographic angiography; ICA, invasive coronary angiography. Agreement is presented as kappa values (95% confidence intervals). * The following variables were not included for comparison given their very low prevalence: severe vessel tortuosity and side branch angulation (n = 4 cases for each).
confidence interval 0.52–0.72). Mean per-patient SYNTAX score derived from ICA was lower in comparison to mean CCTA-derived per-patient SYNTAX score (10.2 ± 8.0 vs 10.9 ± 8.3, P = 0.001). The SYNTAX tertiles for CCTA and ICA, on a per-patient basis, were coincidentally identical and defined as low-SYNTAX ≤ 7, 7 < intermediate-SYNTAX ≤ 12, and high-SYNTAX > 12. Correlation and agreement between CCTA- and ICA-derived SYNTAX
TABLE 2. Reproducibility Analyses—Inter-Observer Variability Between ICA Readers, CCTA Readers, as Well as Intra-observer Variability for Each of the CCTA Readers SYNTAX Score Tertiles* Inter-observer variability ICA CCTA Intra-observer variability CCTA 1st reader CCTA 2nd reader
Number of Obstructive Lesions
Bifurcation Lesions
Total Occlusions
0.81 (0.71–0.92) 0.85 (0.76–0.94)
0.94 (0.91–0.97) 0.95 (0.93–0.96)
0.64 (0.55–0.74) 0.71 (0.59–0.84)
0.96 (0.91–1.0) 0.95 (0.9–1.0)
0.85 (0.69–1.0) 0.78 (0.61–0.94)
0.94 (0.90–0.99) 0.98 (0.96–1.0)
0.71 (0.53–0.90) 0.79 (0.63–0.95)
1.0 0.86 (0.67–1.0)
CCTA, coronary computed tomographic angiography; ICA, invasive coronary angiography. * Data presented as weighted kappa values (95% confidence intervals).
4
Academic Radiology, Vol ■, No ■■, ■■ 2016
CCTA-DERIVED SYNTAX SCORE AND COMPLEX PCI
Figure 2. Correlation between CCTA- and ICA-derived SYNTAX scores on a per-segment, per-vessel, and per-patient basis. CCTA, coronary computed tomographic angiography; ICA, invasive coronary angiography.
scores, on per-patient, per-vessel, and per-segment basis, are presented in Figures 2 and 3a, respectively. Comparison Between Patients Referred to PCI vs CABG or Medical Therapy
The therapeutic decision was taken ad hoc by the operator and was based upon visual assessment of stenosis severity or functional assessment by prior nuclear myocardial perfusion imaging or functional flow reserve at the time of diagnostic angiography or intravascular ultrasound assessment. CCTA, ICA, and PCI were performed in 113 patients for 154 lesions in 131 vessels within less than 30 days between CCTA and PCI. Diagnostic ICA and PCI were performed on the same day for the majority of the patients (95%). Visual assessment of stenosis severity was associated with coronary intervention in 130 (84%) obstructive lesions, whereas functional assessment accounted for the remainder of interventions. There were 15 patients referred for CABG for 48 lesions. Diffuse obstructive disease was associated with the decision
to operate in 13 patients, whereas 2 patients underwent concomitant valve replacement and CABG. Within each therapeutic approach group, the per-patient CCTA- and ICA-derived SYNTAX scores, as well as segmental SYNTAX scores, were similar (Table 4). As expected, the per-patient SYNTAX scores in patients who were referred to surgery were significantly higher in comparison to patients that were referred to PCI or medical management. The segmental SYNTAX scores of CABG lesions were higher in comparison to lesions that were designated to medical management but were similar in comparison PCI lesions (Table 4). Secondary Analysis of Patients Undergoing PCI
CCTA- and ICA-derived per-patient and per-vessel SYNTAX scores demonstrated excellent agreement (Fig 3b) and correlation (Pearson correlation coefficients were 0.99, 0.95, 0.9, and 0.93 for left main (LM), left anterior descending (LAD), left circumflex (LCX), and right coronary artery (RCA), respectively; P < 0.05 for all), with CCTA correctly detecting 5
SHALEV ET AL
Academic Radiology, Vol ■, No ■■, ■■ 2016
Figure 3. Bland-Altman plots for agreement for the (a) entire study cohort and (b) patients undergoing PCI.
6
Academic Radiology, Vol ■, No ■■, ■■ 2016
TABLE 4. CCTA- and ICA-derived Per-patient and Segmental SYNTAX Scores According to Therapeutic Approach Per Patient SYNTAX Score
Medical Management (n = 26 Patients)
PCI (n = 113)
CABG (n = 15)
CCTA ICA Segmental SYNTAX score CCTA ICA
6.8 ± 4.1 7.2 ± 5.4 Medical management (n = 78 lesions) 4.9 ± 3.7 5.0 ± 3.8
10.3 ± 6.3 10.1 ± 6.1 PCI (n = 154 lesions) 5.3 ± 3.5 5.5 ± 3.5
23.5 ± 14.6*,† 21.7 ± 12.1*,† CABG (n = 48 lesions) 6.4 ± 4.5* 6.5 ± 4.4*
CCTA, coronary computed tomographic angiography; CABG, coronary artery bypass graft; ICA, invasive coronary angiography; PCI, percutaneous coronary intervention. * P value <0.05 for comparison of CABG vs medical management. † P value <0.05 for comparison of CABG vs PCI.
all the lesions that led to revascularization. On a per patient basis, SYNTAX scores derived from CCTA vs ICA did not differ (10.3 ± 6.3 vs 10.1 ± 6.1, P = 0.3). For patients undergoing PCI, CCTA-, and ICA-derived segmental SYNTAX scores were similar (5.5 ± 3.6 vs 5.5 ± 3.5, P = 0.7). Similarly, for patients with single-lesion PCI, mean CCTA-, and ICAderived segmental SYNTAX scores were similar (5.9 ± 3.6 vs 5.9 ± 3.5, respectively). Total Contrast Volume
Total contrast volume was moderately correlated with CCTAand ICA-derived segmental SYNTAX score tertiles (r = 0.50, P < 0.05 and r = 0.51, P < 0.05, respectively). Distribution of contrast volumes by SYNTAX tertiles is presented in Table 5.
CCTA-DERIVED SYNTAX SCORE AND COMPLEX PCI
The contrast volume administered at each of the per-patient or segmental SYNTAX score tertiles was not significantly different between CCTA- or ICA-derived SYNTAX tertiles. Ten patients had baseline glomerular filtration rate >0 and <60 mL/min and none developed CIN. Overall CIN grades 1 or 2 were observed in three patients and one patient, respectively. CCTA- and ICA-derived segmental SYNTAX score as a function of contrast media volume was similar among patients who developed CIN as compared to patients who did not (270 ± 95 mL vs 190 ± 93 mL, P = 0.2). Total Fluoroscopy Time
Total fluoroscopy time was moderately correlated with CCTAand ICA-derived segmental SYNTAX score tertiles (r = 0.42, P < 0.05 and r = 0.43, P < 0.05, respectively). Distribution of total fluoroscopy time is presented in Table 5. Of note, there was no correlation between operators’ expertise and metrics of contrast volume utilization or fluoroscopy time. Procedure Failure or Complication
PCI was unsuccessful in four totally occluded lesions (3%). The mean CCTA-derived segmental SYNTAX score for lesions with unsuccessful interventions was significantly higher than for lesions with successful interventions (13 ± 8 vs 5.3 ± 3.2, P < 0.001), a finding in accord with ICA-derived segmental SYNTAX scores. Dissection occurred in additional three lesions (2%) with mean CCTA- and ICA-derived segmental SYNTAX scores of 3.0 ± 1.0 and 3.3 ± 1.1, respectively (P = 0.7). DISCUSSION In the present study we assessed the feasibility and accuracy of a CCTA-derived SYNTAX score to predict complex PCI.
TABLE 5. Total Fluoroscopy Time and Contrast Volume According to Total and Segmental SYNTAX Score Tertiles Among PCI Subgroup (n = 113)
Overall SYNTAX score ICA Contrast volume (mL) Total fluoroscopy time (min) CCTA Contrast volume (ml) Total fluoroscopy time (min) Segmental SYNTAX scores ICA Contrast volume (mL) Total fluoroscopy time (min) CCTA Contrast volume (mL) Total fluoroscopy time (min)
1st Tertile
2nd Tertile
3rd Tertile
P Value
153.4 ± 62.7 12.7 ± 9.2
193.0 ± 64.8 16.4 ± 9.8
251.7 ± 139.4 22.1 ± 14.6
<0.001 <0.001
152.7 ± 62.7 12.6 ± 9.0
188.6 ± 63.7 17.0 ± 10.4
251.8 ± 133.6 21.5 ± 14.2
<0.001 <0.001
135.7 ± 43.2 10.6 ± 4.8
184.2 ± 59.2 17 ± 11.6
222.7 ± 127.1 18.7 ± 12.5
<0.01 0.02
144.3 ± 49 11.5 ± 7.9
194.3 ± 68.4 19.2 ± 11.6
215.4 ± 125.5 17.5 ± 12
0.01 0.03
Mean ± SD
Mean ± SD
CCTA, coronary computed tomographic angiography; ICA, invasive coronary angiography; PCI, percutaneous coronary intervention.
7
SHALEV ET AL
Within the study population, we noted generally high concordance and diagnostic accuracy of the CCTA-derived SYNTAX score as compared to an ICA-derived SYNTAX score reference standard. Further, higher CCTA-derived SYNTAX scores were associated with increasing complexity of coronary revascularization, with higher volumes of iodinated contrast, longer fluoroscopy times, and greater use of advanced devices and procedures. The SYNTAX score is a useful ICA score that incorporates anatomical and morphological lesion characteristics that can be used to reliably determine the complexity of CAD (1,8). SYNTAX score is well-validated, highly reproducible (11,12), and predictive of adverse cardiovascular events in a myriad of scenarios concerning PCI in multivessel and left main CAD (13–15). The SYNTAX score provides greater discriminatory ability compared to the modified American College of Cardiology/American Heart Association lesion classification system (13). CCTA has been established as a non-invasive anatomic test that offers high diagnostic performance for detection and exclusion of high-grade coronary stenosis when compared to traditional ICA. Moreover, CCTA allows for direct visualization of additional CAD features beyond stenosis severity, including measures of lesion length, lesion location (eg, at bifurcations of coronary vessels), and a multitude of other variables commonly used in the ICA SYNTAX (16). Several of these findings of CAD extent, severity, location, and morphology manifest important prognostic implications, and are associated with future death and major adverse cardiovascular events across numerous patient populations (17–22). As previously demonstrated by our investigative group, proximal LAD stenosis, coronary plaque distribution, weighted segmental stenosis scores, and number of vessels with ≥50% and ≥70% stenosis strongly predict mortality (17). Papadopoulou et al. have studied the application of CCTA-derived SYNTAX score in a group of 80 patients, and concluded that CCTA-derived score is feasible, reproducible, and comparable to ICA-derived SYNTAX score (23). Our present study is both in accordance with as well as advances these findings and, to our knowledge, represents the largest study to date to assess the reliability and applicability of a CCTA-derived SYNTAX score. CCTA- and ICA-derived SYNTAX scores demonstrated generally high concordance on a per-segment, per-vessel, and perpatient basis, as well as substantial agreement for the majority of the SYNTAX score variables. CCTA and ICA were in excellent agreement for detection of obstructive lesions and total occlusions, but only in moderate agreement for lesion length. The potential reasons for this are manifold. Given the ability of CCTA to define atherosclerosis in three dimensions rather than two, it is possible that multiplanar reconstructions of coronary vessels by CCTA may provide slightly different measures over ICA which may be susceptible to foreshortening and vessel overlap. Although determination of heavy calcification by CCTA was in good agreement with ICA, severe coronary calcification was prominently visualized and noted as a limitation of currentgeneration CCTA for measures of stenosis severity; as such, this 8
Academic Radiology, Vol ■, No ■■, ■■ 2016
may have led to overestimation by CCTA. Blunt stumps and intracoronary thrombi were completely undetected by CCTA. These disparities may stem from the superior spatial and temporal resolution of ICA over CCTA, and may suggest a need for improved methods of CCTA interpretation. Finally, additional elements of the SYNTAX score, ie, tortuous vessels, were rarely encountered in the study population, were therefore not compared between CCTA and ICA, and underscore the need for future larger studies. Perhaps more important than the demonstration that CCTAderived SYNTAX scores are accurate and concordant with ICA-derived scores is the finding that this non-invasive method is associated with PCI complexity. Although prior studies have reported the usefulness of CCTA to identify chronic total occlusions that are amenable to revascularization (24–27), these investigations are limited in number and scope, have not evaluated the totality of SYNTAX CAD variables, and have not systematically examined the utility of CCTA to identify complex PCI across a range of CAD states, as was done in this study. Stähli et al. have reported their experience with applying a CCTA-derived SYNTAX score for assessment of 66 lesions deemed eligible for intervention (28). The CCTA- and ICAderived SYNTAX scores were comparable and predictive of failed PCI, but not for PCI complexity. Our study directly extends these important observations by providing a comprehensive assessment of PCI complexity by an array of metrics that indicate heightened complexity of PCI. These factors may allow for pre-procedural assessment and planning in a more refined manner, and may allow for enhanced physicianpatient interactions for maximizing informed consent. As complex CAD is often accompanied by baseline chronic kidney disease, concerns regarding a heightened incidence of CIN exist. Further, radiation dose is an important consideration even within an aging population, and the stochastic risk of radiation-induced cancer remains of non-negligible concern. Complex procedures entail higher contrast volume and radiation exposure, as has been reported in an array of prior studies (29) and corroborated by the present study findings. In the present study, CCTA- and ICA-derived segmental SYNTAX scores correlated well between patients with and without CIN. This finding may be largely explained by the relatively low CIN incidence and thus, whether SYNTAX score can predict development of CIN in larger cohorts remains to be determined. This study is not without limitations. While all patients undergoing CCTA were prospectively identified and enrolled, examination of the ICA and PCI of these generally lowintermediate risk patients was performed in a retrospective fashion, and potentially affected by selection or referral bias. Moreover, the study population represented low-intermediate risk individuals who underwent ICA based upon CCTA and thus, whether these findings are applicable to higher risk patients requires further study. As CCTA is prone to overestimation of stenosis severity, the overall accuracy of the CCTA-derived scores may be reduced (3–5,30,31). Further, prior fractional flow reserve
Academic Radiology, Vol ■, No ■■, ■■ 2016
studies have proven the prognostic utility of a “functional” SYNTAX score, and this was not addressed in the current investigation (32). As radiation dose by ICA is difficult to quantify, we relied upon estimates based upon total fluoroscopy time as a surrogate for total effective radiation dose. Finally, as PCIs are primarily performed in an ad hoc manner by interventionalists of potentially different skill levels for patients with various comorbidities, CCTA-based SYNTAX scores cannot alone serve as the definitive guide for prediction of individuals who may experience complex coronary interventions. In conclusion, SYNTAX scores derived from CCTA are concordant with those derived from ICA, and correspond well with complex PCI.
REFERENCES 1. Sianos G, Morel M-A, Kappetein AP, et al. The SYNTAX Score: an angiographic tool grading the complexity of coronary artery disease. EuroIntervention 2005; 1:219–227. 2. Raff GL, Gallagher MJ, Oein O, et al. Diagnostic accuracy of noninvasive coronary angiography using 64-slice spiral computed tomography. J Am Coll Cardiol 2005; 46:552–557. 3. Miller JM, Rochitte CE, Dewey M, et al. Diagnostic performance of coronary angiography by 64-row CT. N Engl J Med 2008; 359:2324–2336. 4. Budoff MJ, Dowe D, Jollis JG, et al. Diagnostic performance of 64multidetector row coronary computed tomographic angiography for evaluation of coronary artery stenosis in individuals without known coronary artery disease: results from the prospective multicenter ACCURACY (Assessment by Coronary Computed Tomographic Angiography of Individuals Undergoing Invasive Coronary Angiography) trial. J Am Coll Cardiol 2008; 52:1724–1732. 5. Meijboom WB, Meijs MFL, Schuijf JD, et al. Diagnostic accuracy of 64slice computed tomography coronary angiography: a prospective, multicenter, multivendor study. J Am Coll Cardiol 2008; 52:2135–2144. 6. Ferencik M, Ropers D, Abbara S, et al. Diagnostic accuracy of image postprocessing methods for the detection of coronary artery stenoses by using multidetector CT. Radiology 2007; 243:696–702. 7. Cheng V, Gutstein A, Wolak A, et al. Moving beyond binary grading of coronary arterial stenoses on coronary computed tomographic angiography: insights for the imager and referring clinician. JACC Cardiovasc Imaging 2008; 1:460–471. 8. Leaman DM, Brower RW, Meester GT, et al. Coronary artery atherosclerosis: severity of the disease, severity of angina pectoris and compromised left ventricular function. Circulation 1981; 63:285–299. 9. Levey AS, Stevens LA, Schmid CH, et al. A new equation to estimate glomerular filtration rate. Ann Intern Med 2009; 150:604–612. 10. Harjai KJ, Raizada A, Shenoy C, et al. A comparison of contemporary definitions of contrast nephropathy in patients undergoing percutaneous coronary intervention and a proposal for a novel nephropathy grading system. Am J Cardiol 2008; 101:812–819. 11. Serruys PW, Onuma Y, Garg S, et al. Assessment of the SYNTAX score in the Syntax study. EuroIntervention 2009; 5:50–56. 12. Garg S, Girasis C, Sarno G, et al. The SYNTAX score revisited: a reassessment of the SYNTAX score reproducibility. Catheter Cardiovasc Interv 2010; 75:946–952. 13. Valgimigli M, Serruys PW, Tsuchida K, et al. Cyphering the complexity of coronary artery disease using the syntax score to predict clinical outcome in patients with three-vessel lumen obstruction undergoing percutaneous coronary intervention. Am J Cardiol 2007; 99:1072–1081. 14. Serruys PW, Morice M-C, Kappetein AP, et al. Percutaneous coronary intervention versus coronary-artery bypass grafting for severe coronary artery disease. N Engl J Med 2009; 360:961–972. 15. Wykrzykowska JJ, Garg S, Girasis C, et al. Value of the SYNTAX score for risk assessment in the all-comers population of the randomized multicenter LEADERS (Limus Eluted from a durable versus ERodable stent coating) trial. J Am Coll Cardiol 2010; 56:272–277.
CCTA-DERIVED SYNTAX SCORE AND COMPLEX PCI
16. Voros S, Rinehart S, Qian Z, et al. Coronary atherosclerosis imaging by coronary CT angiography: current status, correlation with intravascular interrogation and meta-analysis. JACC Cardiovasc Imaging 2011; 4:537– 548. 17. Min JK, Shaw LJ, Devereux RB, et al. Prognostic value of multidetector coronary computed tomographic angiography for prediction of allcause mortality. J Am Coll Cardiol 2007; 50:1161–1170. 18. Hulten EA, Carbonaro S, Petrillo SP, et al. Prognostic value of cardiac computed tomography angiography: a systematic review and metaanalysis. J Am Coll Cardiol 2011; 57:1237–1247. 19. Min JK, Dunning A, Lin FY, et al. Age- and sex-related differences in allcause mortality risk based on coronary computed tomography angiography findings results from the International Multicenter CONFIRM (COroNary CT Angiography Evaluation For Clinical Outcomes: An InteRnational Multicenter Registry) of 23,854 patients without known coronary artery disease. J Am Coll Cardiol 2011; 58:849–860. 20. Rana JS, Dunning A, Achenbach S, et al. Differences in prevalence, extent, severity, and prognosis of coronary artery disease among patients with and without diabetes undergoing coronary computed tomography angiography: results from 10,110 individuals from the CONFIRM (COroNary CT Angiography Evaluation For Clinical Outcomes: An InteRnational Multicenter Registry). Diabetes Care 2012; 35:1787–1794. 21. Chow BJW, Small G, Yam Y, et al. Incremental prognostic value of cardiac computed tomography in coronary artery disease using CONFIRM: COroNary CT Angiography Evaluation For Clinical Outcomes: An InteRnational Multicenter Registry. Circ Cardiovasc Imaging 2011; 4:463– 472. 22. Villines TC, Hulten EA, Shaw LJ, et al. Prevalence and severity of coronary artery disease and adverse events among symptomatic patients with coronary artery calcification scores of zero undergoing coronary computed tomography angiography: results from the CONFIRM (COroNary CT Angiography Evaluation For Clinical Outcomes: An InteRnational Multicenter Registry). J Am Coll Cardiol 2011; 58:2533–2540. 23. Papadopoulou S-L, Girasis C, Dharampal A, et al. CT-SYNTAX score: a feasibility and reproducibility study. JACC Cardiovasc Imaging 2013; 6:413– 415. 24. Mollet NR, Hoye A, Lemos PA, et al. Value of preprocedure multislice computed tomographic coronary angiography to predict the outcome of percutaneous recanalization of chronic total occlusions. Am J Cardiol 2005; 95:240–243. 25. Soon KH, Cox N, Wong A, et al. CT coronary angiography predicts the outcome of percutaneous coronary intervention of chronic total occlusion. J Interv Cardiol 2007; 20:359–366. 26. García-García HM, van Mieghem CAG, Gonzalo N, et al. Computed tomography in total coronary occlusions (CTTO registry): radiation exposure and predictors of successful percutaneous intervention. EuroIntervention 2009; 4:607–616. 27. Hsu JT, Kyo E, Chu CM, et al. Impact of calcification length ratio on the intervention for chronic total occlusions. Int J Cardiol 2011; 150:135– 141. 28. Stähli BE, Bonassin F, Goetti R, et al. Coronary computed tomography angiography indicates complexity of percutaneous coronary interventions. J Invasive Cardiol 2012; 24:196–201. 29. Wertman BM, Cheng VY, Kar S, et al. Characterization of complex coronary artery stenosis morphology by coronary computed tomographic angiography. JACC Cardiovasc Imaging 2009; 2:950–958. 30. Meijboom WB, Van Mieghem CAG, van Pelt N, et al. Comprehensive assessment of coronary artery stenoses: computed tomography coronary angiography versus conventional coronary angiography and correlation with fractional flow reserve in patients with stable angina. J Am Coll Cardiol 2008; 52:636–643. 31. Schuijf JD, Wijns W, Jukema JW, et al. Relationship between noninvasive coronary angiography with multi-slice computed tomography and myocardial perfusion imaging. J Am Coll Cardiol 2006; 48:2508–2514. 32. Nam C-W, Mangiacapra F, Entjes R, et al. Functional SYNTAX score for risk assessment in multivessel coronary artery disease. J Am Coll Cardiol 2011; 58:1211–1218.
SUPPLEMENTARY DATA Supplementary data related to this article can be found at http://dx.doi.org/10.1016/j.acra.2016.07.003.
9