Article Text

Download PDFPDF

Widening clinical applications of the SYNTAX Score
  1. Vasim Farooq1,
  2. Stuart J Head2,
  3. Arie Pieter Kappetein2,
  4. Patrick W Serruys1
  1. 1Department of Interventional Cardiology, Erasmus University Medical Centre, Thoraxcenter, Rotterdam, The Netherlands
  2. 2Department of Cardiothoracic Surgery, Erasmus University Medical Centre, Thoraxcenter, Rotterdam, The Netherlands
  1. Correspondence to Professor Patrick W Serruys, Department of Interventional Cardiology, Erasmus MC, 's-Gravendijkwal 230, Rotterdam 3015 CE, The Netherlands; p.w.j.c.serruys{at}erasmusmc.nl

Abstract

The SYNTAX Score (http://www.syntaxscore.com) has established itself as an anatomical based tool for objectively determining the complexity of coronary artery disease and guiding decision-making between coronary artery bypass graft (CABG) surgery and percutaneous coronary intervention (PCI). Since the landmark SYNTAX (Synergy between PCI with Taxus and Cardiac Surgery) Trial comparing CABG with PCI in patients with complex coronary artery disease (unprotected left main or de novo three vessel disease), numerous validation studies have confirmed the clinical validity of the SYNTAX Score for identifying higher-risk subjects and aiding decision-making between CABG and PCI in a broad range of patient types. The SYNTAX Score is now advocated in both the European and US revascularisation guidelines for decision-making between CABG and PCI as part of a SYNTAX-pioneered heart team approach. Since establishment of the SYNTAX Score, widening clinical applications of this clinical tool have emerged. The purpose of this review is to systematically examine the widening applications of tools based on the SYNTAX Score: (1) by improving the diagnostic accuracy of the SYNTAX Score by adding a functional assessment of lesions; (2) through amalgamation of the anatomical SYNTAX Score with clinical variables to enhance decision-making between CABG and PCI, culminating in the development and validation of the SYNTAX Score II, in which objective and tailored decisions can be made for the individual patient; (3) through assessment of completeness of revascularisation using the residual and post-CABG SYNTAX Scores for PCI and CABG patients, respectively. Finally, the future direction of the SYNTAX Score is covered through discussion of the ongoing development of a non-invasive, functional SYNTAX Score and review of current and planned clinical trials.

  • CARDIAC SURGERY
  • CORONARY ARTERY DISEASE

Statistics from Altmetric.com

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Introduction

The SYNTAX Score (http://www.syntaxscore.com) has emerged as an anatomically based tool for objectively determining the complexity of coronary artery disease and guiding decision-making between coronary artery bypass graft (CABG) surgery and percutaneous coronary intervention (PCI).1–4 Since the landmark SYNTAX (Synergy between PCI with Taxus and Cardiac Surgery) Trial5–7 comparing CABG with PCI in patients with complex coronary artery disease (unprotected left main coronary artery (ULMCA) or de novo three vessel disease (3VD)), numerous validation studies have confirmed the clinical validity of the SYNTAX Score for identifying higher-risk subjects and aiding decision-making between CABG and PCI in a broad range of patient types.4 ,8 The SYNTAX Score is now advocated in both the European and US revascularisation guidelines9–11 as part of the SYNTAX-pioneered heart team approach.12 In addition, the US Food and Drug Administration mandates the SYNTAX Score as an entry criterion in ongoing contemporary stent and structural heart disease trials—namely, the EXCEL (Evaluation of XIENCE PRIME or XIENCE V Everolimus Eluting Stent System Versus Coronary Artery Bypass Surgery for Effectiveness of Left Main Revascularisation) Trial (ClinicalTrials.gov identifier: NCT01205776)13 and SURTAVI (Safety and Efficacy Study of the Medtronic CoreValve System in the Treatment of Severe, Symptomatic Aortic Stenosis in Intermediate Risk Subjects Who Need Aortic Valve Replacement) Trial (ClinicalTrials.gov identifier: NCT01586910).

Since establishment of the SYNTAX Score, widening clinical applications of this clinical tool have emerged (table 1). The purpose of this review is to give the clinician a concise overview of the widening applications of the SYNTAX Score, from improving its diagnostic accuracy by incorporation of a functional component, augmenting the anatomical SYNTAX Score with clinical variables to enhance decision-making between CABG and PCI, moving towards individualised decision-making between CABG and PCI, to assessment of the completeness of revascularisation and its prognostic implications. Lastly, the future direction of the SYNTAX Score is explored.

Table 1

Outline of the SYNTAX Score and how it has progressed since its inception

Anatomical and functional approach

SYNTAX Score

The SYNTAX Score was developed during the design of the SYNTAX Trial as a tool to force the interventional cardiologist and cardiac surgeon to systematically analyse the coronary angiogram and to specify the number of coronary lesions requiring treatment, their angiographic location and anatomical complexity.1–5 ,8 The SYNTAX Score combines the importance of a diseased coronary artery segment in terms of its severity (ie, obstructive or occlusive), anatomical location and importance in supplying blood to the myocardium (‘vessel-segment weighting’ based on the Leaman Score14), adverse lesion characteristics (American College of Cardiology (ACC)/American Heart Association (AHA) lesion classification),15 bifurcation lesion characteristics (Medina classification16) and total occlusion characteristics from the European TOTAL Surveillance Study.17 Each vessel segment, 1.5 mm in diameter or greater (figure 1A, labelled 1–16), with a ≥50% diameter stenosis by visual estimation, is awarded a multiplication factor related to coronary lesion location and severity (figure 1A). Further characterisation of the coronary lesions leads to the addition of more points (figure 1B), which includes features of total occlusions (duration, length, blunt stump, presence of bridging collaterals or side branch), bifurcation (Medina classification16) or trifurcation (number of diseased branches involved), side branch angulation, aorto-ostial lesion, severe tortuosity, lesion length >20 mm, heavy calcification, thrombus, and diffuse or small vessel disease. An online SYNTAX Score algorithm1 automatically summates each of these features to calculate the total SYNTAX Score.

Figure 1

Coronary tree segments and their importance in supplying blood flow to the left ventricle (vessel segment weighting—weighting factors—Leaman Score14) based on the presence of a right or left dominant system (A). A multiplication factor of ×2 is used for non-occlusive (50–99% diameter stenosis) lesions and ×5 for occlusive (100% diameter stenosis) lesions. For example, a stenotic proximal LAD lesion (segment 6) would have a weighting factor of 3.5×2 (7 points), and an occlusive proximal LAD lesion a weighting factor of 3.5×5 (17.5 points). Other adverse lesion characteristics considered in the SYNTAX score have an additive value (B). Images used with permission from the SYNTAX Trial Investigators.

Based primarily on the results of the SYNTAX Trial,5–7 current European revascularisation guidelines9 give subjects with 3VD and low SYNTAX Scores (0–22) a level of evidence of IA for CABG and IIa B for PCI. In subjects with ULMCA disease and low to intermediate SYNTAX Scores (<33), a level of evidence of IA is given for CABG and IIb B for PCI. Furthermore, US guidelines now give surgical revascularisation for ULMCA disease a Class 1B recommendation,10 ,11 compared with a Class 1A recommendation in previous guidelines.18

Functional SYNTAX Score

PCI guided by the assessment of the functional significance of a lesion using fractional flow reserve (FFR) has been shown to improve clinical outcomes.19 The functional SYNTAX Score uses the principle of the functional assessment of coronary lesions to determine the SYNTAX Score, rather than the angiographic determination of the SYNTAX Score based on visual assessment, as is undertaken in conventional SYNTAX Score calculations. In a retrospective sub-analysis of almost 500 patients (n=497) from the FFR-guided arm of the FAME (Fractional Flow Reserve vs Angiography for Multivessel Evaluation) Study, the primary benefit was reclassifying higher-risk groups into lower-risk categories without any adverse sequelae in terms of major adverse cardiac events (MACE) and death or myocardial infarction (MI) at 1 year.20

It should be emphasised that subjects in the FAME Study had substantially less complex coronary artery disease (mean±SD SYNTAX Score 14.8±6.0) compared with the PCI arm of the SYNTAX Trial (mean±SD SYNTAX Score 28.4±11.5), and that subjects with left main coronary artery disease were not investigated. Prospective validation studies of the functional SYNTAX Score in complex coronary artery disease are awaited at the time of writing.

Augmenting the anatomical SYNTAX Score with clinical factors and the personalisation of decision-making: development of the SYNTAX Score II

Since the SYNTAX Score was developed, limitations of this scoring system in aiding decision-making between CABG and PCI has become evident—namely, the lack of clinical variables and lack of a personalised approach to decision-making. Below is a brief overview of the ‘development phase’ leading to the SYNTAX Score II, which was designed to overcome these limitations (table 1).

The law of ‘parsimony’ and ‘ACEF’

ACEF

Ranucci et al21–23 developed a simple risk model consisting of only three clinical variables (age, serum creatinine and left ventricular ejection fraction (LVEF)), for assessing operative mortality risk in elective cardiac operations (ACEF Score—figure 2A). Based on the ‘law of parsimony’ or ‘the Ockham razor’ concept, whereby a simple model can explain a phenomenon with the same level of accuracy as complex models, ACEF was shown to be least comparable to the EuroSCORE (composed of 17 variables)24 ,25 in predicting in-hospital mortality after CABG.22 ,23

Figure 2

Side by side comparisons of ACEF (A) and the Logistic Clinical SYNTAX Score (B). Images used with permission from Ranucci et al23 and Farooq et al.32 *SYNTAX-like patient defined as fulfilling the enrolment criteria for the SYNTAX All-Comers trial, i.e. left main stem (isolated or associated with one-, two-, or three-vessel disease) or three-vessel disease alone. CrCl, creatinine clearance, LV ejection fraction, left ventricular ejection fraction.

The three risk factors used in ACEF are natural continuous variables that are objectively defined and not subject to personal estimation (eg, is the patient diabetic? does the patient have extra cardiac arteriopathy?). In addition, the variables of ACEF are known independent risk factors for mortality, and it was subsequently shown that the end organ manifestations of the risk factor (as identified in ACEF) are more important than the actual presence of the risk factor for predicting long-term prognosis.21 ,26–28

Clinical SYNTAX Score/Logistic Clinical SYNTAX Score

Based on the principle of ACEF, the Clinical SYNTAX Score,29–31 and subsequently the Logistic Clinical SYNTAX Score (figure 2B),32 ,33 were developed and validated. Both the Clinical SYNTAX Score and Logistic Clinical SYNTAX Score combined ACEF with the SYNTAX Score, and were shown to improve mortality predictions compared with the SYNTAX Score alone in subjects with complex coronary artery disease.29–33 Similar to the conventional SYNTAX Score, the Clinical SYNTAX Score relied on categorisation of risk (low, intermediate and high) and was able to only identify a high-risk group after PCI.2931 The Logistic Clinical SYNTAX Score was designed to individualise risk and provide 1-year mortality predictions in an all-comers PCI population irrespective of clinical presentation (except cardiogenic shock).32 ,33

The Logistic Clinical SYNTAX Score was developed and cross-validated (‘internal–external’ validation procedure34) in >6000 subjects from seven contemporary coronary stent trials,32 and further externally validated in 2627 subjects presenting with non-ST elevation acute coronary syndrome and undergoing PCI from the Acute Catheterisation and Urgent Intervention Triage Strategy (ACUITY) Trial.33 Notably, the addition of a further six additional clinical variables, including diabetes, to the Logistic Clinical SYNTAX Score led to only a minor incremental improvement in risk predictions.32 ,33 Thus, the Logistic Clinical SYNTAX Score was shown to follow the law of parsimony, as seen with the surgical ACEF model discussed above2123—namely, the end organ manifestations of the risk factor were more important than the actual presence of the risk factor for predicting long term prognosis.

Global risk

In the SYNTAX Trial, it was shown that, as well as the SYNTAX Score in PCI subjects, the EuroSCORE (a surgery-based risk score composed of 17 variables designed to predict in-hospital mortality after CABG24 ,25) was an independent predictor of MACE in subjects undergoing surgery- or percutaneous-based revascularisation. Subsequently, it was hypothesised that the amalgamation of the SYNTAX Score with the EuroSCORE could improve decision-making between CABG and PCI.4 ,5 The feasibility of this ‘Global Risk’ approach was demonstrated in a registry of 255 subjects with left main coronary artery disease using tertiles of the SYNTAX Score and tertiles of the additive EuroSCORE that reflected their study population.35 Subsequently, the Global Risk was validated in the SYNTAX Trial using conventional tertiles of the SYNTAX Score and EuroSCORE,36 and was shown to substantially enhance the identification of low-risk patients with ULMCA disease or de novo 3VD who could safely and efficaciously be treated with CABG or PCI, compared with the SYNTAX Score alone.

One of the unexpected findings from the Global Risk was that higher-risk subjects (high additive EuroSCORE ≥6) in all tertiles of the SYNTAX Score (low, intermediate or high) were shown to have a potential prognostic benefit from undergoing CABG compared with PCI, irrespective of the SYNTAX Score, provided that an acceptable threshold of operative risk was not exceeded.36 For example, in the 3VD cohort of the SYNTAX Trial, the 3-year mortality of subjects with a low SYNTAX Score (<23) and a high EuroSCORE (≥6) was doubled when undergoing PCI (15.9%) compared with CABG (8.2%). One hypothesis used to explain these findings is that the bypass graft would potentially ‘protect’ the entire treated coronary vessel from future cardiac events for the lifespan of the graft in high-risk subjects compared with PCI, which would treat the individual lesion.37 On the basis of these observations, it was hypothesised by the investigators that potentially low (or high) risk subjects were potentially concealed by high (or low) risk subjects in all tertiles of the SYNTAX Score. This hypothesis is what prompted the investigators to develop a more individualised approach to decision-making between CABG and PCI, and subsequently led to the development of the SYNTAX Score II,38 as detailed below.

SYNTAX Score II

As previously discussed, the combination of the anatomical SYNTAX Score with ACEF contained most of the prognostic information for predicting mortality after CABG (excluding the anatomical SYNTAX Score2123 ,36) or PCI (including the anatomical SYNTAX Score32 ,36). The SYNTAX Score II was built on the ACEF ‘skeleton’, with the addition of risk factors that were shown to directly affect decision-making between CABG and PCI—that is, interaction effects, namely a risk factor being more predictive of mortality in patients undergoing PCI compared with CABG, or vice versa (figure 3).38 For example, the anatomical SYNTAX Score aids decision-making between CABG and PCI because it is more predictive of clinical outcomes in patients undergoing PCI than in patients undergoing CABG (where it is not predictive). On the basis of this principle, younger age, female gender and reduced LVEF favoured CABG over PCI on long-term prognostic grounds. Thus, in such patients, a lower anatomical SYNTAX Score would be required in order for the long-term mortality risk to be similar between CABG and PCI. In contrast, older age, chronic obstructive pulmonary disease (COPD) or ULMCA disease favoured PCI over CABG, and thus, in this type of patient, a higher anatomical SYNTAX Score would be needed for the long-term mortality risks to be similar.

Figure 3

Predictor effects for coronary artery bypass graft (CABG) and percutaneous coronary intervention (PCI) in the SYNTAX Score II. These are represented visually as log HR for CABG and PCI on the y-axis for each predictor. Each predictor is expressed on the x-axis continuously (upper) or categorically (lower), for a person of mean baseline characteristics. Diabetes is included (highlighted in red) to illustrate its absence of interaction when included in the analyses. Note the different gradients of the hazards for PCI and CABG, leading to the hazards crossing at an anatomical SYNTAX Score of 15. At this ‘cross-over’ point of hazards, the mortality risk is comparable between CABG and PCI. This threshold of cross-over of hazards will vary according to the level of other variables, namely being lower for female gender, reduced left ventricular ejection fraction (LVEF) and younger age, and higher for chronic obstructive pulmonary disease (COPD), unprotected left main coronary artery disease and older age. As both PVD (p=1.00) and diabetes (p=0.67) lacked an interaction effect, as indicated by almost parallel HRs (ie, comparable increase in mortality risk), their presence would have no effect on decision-making between CABG and PCI. Legend and image reproduced with permission from Farooq et al.3 ,38 3VD three vessel disease; LMS left main stem; CrCl creatinine clearance; COPD chronic obstructive pulmonary disease; PVD peripheral vascular disease.

By adopting the individualised approach of the SYNTAX Score II, augmented by clinical variables, it was shown that subsets of patients existed in all tertiles of the SYNTAX Score in which CABG or PCI would confer a mortality benefit, or offer similar long-term prognosis.38 A nomogram was developed (figure 4) that allowed an accurate individualised prediction of 4-year mortality in patients proposing to undergo CABG or PCI, to objectively aid decision-making. For example, a 60-year-old man with an anatomical SYNTAX Score of 30, ULMCA disease, CrCl of 60 mL/min, a LVEF of 50%, and COPD would have 41 points (predicted 4-year mortality 16.3%) and 33 points (predicted 4-year mortality 8.7%) for CABG and PCI, respectively. The same example, without COPD included, would lead to identical points (29 points) and identical 4-year mortality predictions (6.3%) for CABG and PCI. An online version of the SYNTAX Score II will soon appear alongside the original SYNTAX Score calculator (http://www.syntaxscore.com).

Figure 4

The SYNTAX Score II nomogram for bedside application. The total number of points for eight factors can be used to accurately predict 4-year mortality for the individual patient proposing to undergo coronary artery bypass graft (CABG) or percutaneous coronary intervention (PCI). Image reproduced with permission from Farooq et al.38 *Due to the rarity of complex coronary artery disease in pre-menopausal women, mortality predictions in younger women are predominantly based on the linear relationship of age with mortality. The differences in mortality predictions in younger women between CABG and PCI will therefore be affected by larger 95% confidence intervals than those in older women but will be equally valid. 3VD three vessel disease; LMS left main stem; CrCl creatinine clearance; LVEF left ventricular ejection fraction; COPD chronic obstructive pulmonary disease; PVD peripheral vascular disease.

Diabetics

Notably, diabetes was not included in the final SYNTAX Score II, despite medically treated diabetes being prestratified at randomisation as a powered subgroup in the SYNTAX Trial, and present in over a quarter of the study patients (26%). This in spite of diabetics being perceived as a specific high-risk group potentially warranting a different treatment strategy compared with patients with other risk factors.911 ,39

The primary reason for the non-inclusion of diabetes in the SYNTAX Score II was that it was shown to be equally predictive of mortality in the CABG and PCI cohorts of the SYNATX Trial, after adjustment for other risk factors (figure 3) —that is, diabetes lacked an interaction effect and was therefore not important for decision-making between CABG and PCI (after adjustment for the end organ manifestations of the risk factors). As previously discussed for ACEF, the end organ manifestation of diabetes was what affected long-term mortality in CABG and PCI populations2123 ,28 ,32—and therefore decision-making between CABG and PCI in the SYNTAX Score II—and not the actual presence of the risk factor. The findings of the lack of inclusion of diabetes in the SYNTAX Score II are supported by epidemiological data, in which non-diabetics with chronic kidney disease and proteinuria had a stronger association with the risk of MI, and a higher mortality, than diabetics, and that the relative risk of long-term mortality associated with chronic kidney disease was ‘much the same irrespective of the presence or absence of diabetes’.26 ,27

Validation of the SYNTAX Score II

As compared with existing revascularisation guidelines using the anatomical SYNTAX Score,911 it was shown that, if CABG or PCI was selected on the basis of a higher or lower expected survival (irrespective of the margin of difference) with the SYNTAX Score II in the SYNTAX Trial, the SYNTAX Score II would only need to be used in 110 patients to have one more patient alive at 4 years.40

External validation of the SYNTAX Score II38 was performed in the multinational Drug Eluting stent for Left main coronary Artery disease (DELTA) Registry (14 centres in Europe, USA and South Korea),41 composed of subjects with ULMCA disease associated with or without multivessel disease (26% of the study population had 3VD). All variables in the SYNTAX Score II interacted in a similar way, and therefore influenced decision-making between CABG and PCI, in the SYNTAX Trial and DELTA Registry, with the exception of age and LVEF, which had minimal interactions in the DELTA Registry—findings that may relate to the unavoidable selection bias inherent to all registries, since decision-making between CABG and PCI has already been made and would be difficult to control for.42 Even randomised trials lacking an all-comers design, with restrictive inclusion and exclusion criteria, can potentially make application to clinical practice questionable.37 ,43 ,44 This is exemplified in a recent meta-analysis of randomised trials undertaken before SYNTAX comparing PCI with CABG, where in most trials 2–12% of screened subjects were randomised because of the highly restrictive inclusion and exclusion criteria.45 In this meta-analysis, CABG was shown to be favoured in older subjects, and PCI in younger subjects45—findings that have since been directly contradicted by the SYNTAX Score II in the all-comers SYNTAX Trial (where the opposite was shown).38 Hence ‘randomised’ validation of the SYNTAX Score II was proposed,38 in which its further validation would be conducted in randomised controlled trials or prospectively run studies, as discussed below under Future directions.

Tools for assessment of completeness of revascularisation

Interpreting the long-term prognostic impact of incomplete revascularisation in patients with complex coronary artery disease has historically been difficult.46 The lack of standardised definitions of incomplete revascularisation has confounded this issue and made comparisons between studies difficult. The residual and post-CABG SYNTAX Score were designed to overcome this limitation as detailed.

Residual SYNTAX score

The residual SYNTAX Score is based on the principle of being a measure of the myocardial ischaemia burden, dependent on the location of the coronary disease, its importance in supplying blood to the myocardium, and the anatomical complexity (eg, calcification, bifurcation, long lesion) associated with the obstructive disease. The residual SYNTAX Score is essentially the SYNTAX Score recalculated after the PCI procedure, and provides an objective, quantitative measure of the degree and complexity of residual stenosis after revascularisation. More proximal coronary artery disease scores more highly on the residual SYNTAX Score since this is dependent on the vessel-segment weighting as previously discussed (figure 1), particularly if the obstructive disease is more complex.47 ,48

Genereux et al47 first demonstrated that a residual SYNTAX Score of >8.0 after PCI was associated with adverse 1-year mortality, in a post hoc analysis of the ACUITY Trial. The ACUITY Trial consisted of subjects with moderate- to high-risk acute coronary syndrome undergoing PCI, and substantially less complex coronary artery disease (median SYNTAX Score 9.0, IQR 5.0–16.0) compared with the SYNTAX Trial (median SYNTAX Score 28, IQR 20.0–36.0).

The residual SYNTAX Score was subsequently validated in the randomised, all-comers SYNTAX Trial, consisting of subjects with complex coronary artery disease (ULMCA or de novo 3VD) at the final 5-year follow-up.48 The previous finding of a residual SYNTAX Score of >8 being associated with adverse long-term clinical outcomes in the ACUITY Trial47 was found to be equally of importance in SYNTAX patients who underwent a 5-year follow-up. Notably, as the baseline SYNTAX Score increased, the frequency of a residual SYNTAX Score >8 increased in unison, with an associated increase in long-term mortality (figure 5). In addition, progressively higher residual SYNTAX Scores were shown to be a surrogate marker of sicker patients,49 with greater baseline clinical comorbidity and anatomical complexity, with consequent adverse long-term clinical outcomes, including all-cause mortality.

Figure 5

The residual SYNTAX Score in the SYNTAX Trial. Complete (residual SYNTAX Score 0) and incomplete (tertiles of the residual SYNTAX Score (residual SYNTAX Score >0)) revascularisation, stratified according to tertiles of the baseline SYNTAX Score (A). Kaplan–Meier curves showing cumulative event rates through to 5 years, based on complete (residual SYNTAX Score 0) and incomplete (tertiles of residual SYNTAX Score) revascularisation, in the low (0–22), intermediate (23–32) and high (≥33) baseline SYNTAX Scores (B). Note the progressive increase in the frequency of a residual SYNTAX Score >8 across the tertiles of the baseline SYNTAX Score (A) and its association with adverse long-term mortality (B). Legend and image reproduced with permission from Farooq et al.48

Stratified analyses in the powered subgroups of ULMCA disease and medically treated diabetes showed a residual SYNTAX Score of >8 to be associated with adverse long-term clinical outcomes, including mortality. Stratified analyses in subjects with reduced LVEF also showed the results to be equally applicable, whereas, in subjects with total occlusions, a more modest effect was shown that did not reach statistical significance. The latter perhaps implied that appropriate viability assessment was required to ensure that revascularisation of the total occlusion was appropriate and clinically justified.50

In summary, the residual SYNTAX Score allowed the quantification of the degree of revascularisation, and the determination of an objective level of reasonable incomplete revascularisation,46 whereby a threshold value could be determined (≤8) that would not have a negative effect on long-term mortality and other clinical outcomes.

Post-CABG SYNTAX Score

The CABG equivalent of the residual SYNTAX Score—post CABG SYNTAX Score—has recently been shown to be linked to adverse 5-year clinical outcomes, including mortality, in the angiographic substudy of the SYNTAX Trial (SYNTAX-LE MANS) (figure 6A).51 ,52 Owing to the inherently different mechanisms of treatment of coronary artery disease with CABG and PCI, calculation of the residual SYNTAX Score (ie, burden of coronary disease removed by PCI) differs from that of the post-CABG SYNTAX Score (ie, coronary disease bypassed with a graft). The basic principle of the post-CABG SYNTAX Score is that it deducts points from the ‘native’ baseline SYNTAX Score based on the level of ‘protection’ conferred by the bypass grafts, through deduction of the vessel-segment weighting (Leaman Score14figure 1) that the bypass graft provides (figure 6B). Since the post-CABG SYNTAX Score is based on validated physiological principles of blood flow (Leaman Score14), it does not arbitrarily deduct points for the type of bypass graft anastomosed. Points related to native coronary disease (eg, bifurcation disease, calcification, total occlusions, long lesions, diffuse disease) remain unaltered.

Figure 6

(A) The post-coronary artery bypass graft (CABG) SYNTAX Score in the angiographic substudy of the SYNTAX Trial (SYNTAX-LE MANS). Outcomes (Kaplan–Meier curves) separated by the median of the post-CABG SYNTAX Score into low (0–21; n=58) and high (≥22; n=55) score groups. At 5 years, significantly greater all-cause mortality, significantly greater all-cause death/CVA/myocardial infarction (MI) and major adverse cardiovascular and cerebrovascular events (MACCE) were evident in the high post-CABG SYNTAX Score group compared with the low post-CABG SYNTAX Score group. Note the peak in MACCE at approximately 18 months secondary to patients undergoing scheduled coronary angiography, the findings of which triggered repeat revascularisation. (B) Example of the calculation of the post-CABG SYNTAX Score. Occluded left main (#) in a left dominant system gave a native SYNTAX Score of 39 (upper image). A patent left internal mammary artery (LIMA) inverse Y graft anastomosed to the mid LAD (upper white arrow), with sequential anastomoses to the 1st, 2nd and 3rd obtuse marginal (OM) branches (lower three white arrows) are shown. Based on the vessel segment weighting (refer to figure 1), 17.5 (occluded proximal LAD) and 12.5 (occluded proximal left circumflex (LCx)) points were deducted from the native SYNTAX Score. Post-CABG SYNTAX Score was therefore 39−17.5−12.5=9 points. Legend and images reproduced with permission from Farooq et al.51 ,52

Historical evidence to back up the findings from the post-CABG SYNTAX Score being linked to adverse long-term prognosis comes from the CASS (Coronary Artery Surgery Study)53 and Rotterdam54 registries. First, in both studies, more extensive preoperative coronary artery disease was linked to the higher prevalence and severity of other clinical risk factors and adverse long-term prognosis, compared with subjects with less complex coronary artery disease. Second, 5 years follow-up of the Bypass Angioplasty Revascularisation Investigation (BARI) Trial demonstrated that native coronary disease progression (and not the extent of initial revascularisation) was the predominant determinant of the recurrence of angina and jeopardised myocardium in percutaneous and surgically revascularised subjects.55 Lastly, coronary artery calcification has been linked to adverse all-cause mortality at 10 years, independent of other risk factors.56 ,57 Validation studies of the post-CABG SYNTAX Score are awaited at the time of writing.

Future directions

Non-invasive SYNTAX Score

Multislice CT SYNTAX Score

Papadopoulou et al58 first described the feasibility and reproducibility of a multislice CT (MSCT)-derived SYNTAX Score in 80 consecutive patients with symptomatic angina, using definitions of the angiographically defined SYNTAX Score adapted for the MSCT capabilities. The underlying concept was to allow the SYNTAX Score to be calculated before the intervention, to potentially aid decision-making and optimise patient management. In this study, the MSCT SYNTAX Score was shown to be feasible, with results comparable to the SYNTAX Score calculated with conventional coronary angiography. While this study was shown to be highly reproducible,58 a subsequent validation study of similar size (n=104) found only fair agreement between MSCT and angiography-derived SYNTAX Scores,59 although this did improve substantially when analyses were restricted to good quality MSCT. Notably, both studies investigated subjects with predominantly less complex coronary artery disease (low SYNTAX Scores <23).58 ,59 Larger scale validation studies, particularly in more complex ‘SYNTAX-like’ patients, are awaited.

Non-invasive FFR

The addition of a non-invasive FFR component (Heartflow, Redwood City, California, USA) has the potential to allow the non-invasive calculation of a functionally based MSCT SYNTAX Score. This technology is based on using computational fluid dynamic techniques applied to the MSCT angiography.58 Validation data from the non-invasive FFR MSCT have been reported in the DISCOVER FLOW and multicentre DeFACTO (Determination of Fractional Flow Reserve by Anatomic Computed Tomographic Angiography) Trials.4 ,60–62 In both studies, non-invasive FFR MSCT was shown to substantially improve the diagnostic accuracy in detecting haemodynamically significant coronary artery disease in subjects with suspected coronary artery disease. Application of this technology in subjects with more complex coronary artery disease has recently been undertaken (figure 7). At the time of writing, work to derive a non-invasive functional MSCT SYNTAX Score using this technology is in progress.

Figure 7

Application of non-invasive fractional flow reserve (FFR) technology during multislice CT (MSCT) in subjects with complex coronary artery disease, namely multivessel (A) and left main and multivessel (B) coronary artery disease. Values shown are the non-invasive calculated FFR in all major epicardial vessels and side branches.

Ongoing and future studies

EXCEL Trial

The ongoing international multicentre EXCEL Trial is aiming to recruit 2600 patients with ULMCA disease and a SYNTAX Score <33, randomised to CABG (n=1300) or PCI with contemporary stents (n=1300).13 The primary end point is a composite measure of all-cause death, MI or stroke 3 years after revascularisation. As part of the EXCEL Trial, validation of the SYNTAX Score II has recently been prespecified as an end point.

SYNTAX II trial

The planned SYNTAX II single-arm trial will use the SYNTAX Score II38 as a tool to recruit subjects with de novo 3VD (without left main involvement) on the grounds of patient safety —that is, subjects with a similar long-term mortality between CABG and PCI, as determined by the SYNTAX Score II in conjunction with the heart team. Notably, subjects from all tertiles of the SYNTAX Score will be eligible. The PCI procedure will be guided by a functional assessment of all three vessels (functional SYNTAX Score20), a newer generation stent platform with a biodegradable polymer,63 ,64 and intravascular ultrasound-guided stent implantation.65 The PCI and CABG arms of the SYNTAX Trial5–7 will act as control arms. Validation of the SYNTAX Score II will be prespecified as an end point. Recruitment of patients is expected to begin towards the end of the year.

Summary

Since the founding of the original SYNTAX Score, a multitude of clinical applications have emerged as discussed in this review. It is, however, important to recognise that randomised trials lacking an all-comers trial design, and registries (however well designed), have their inherent limitations as discussed, and further validation of SYNTAX-based tools in these populations should be interpreted with caution. So-called ‘randomised validation’ may prove to be the better technique in further validating these tools to remove any form of selection bias, particularly in registries, where decision-making between CABG and PCI has already been undertaken. Moreover, the move away from categorisation of risk (ie, low, intermediate or high) to individual risk profiling, culminating in the development of the SYNTAX Score II, is a significant step forward, provided that it is undertaken in the context of the heart team in open dialogue with the patient, whose individual perception of short- and long-term risk would prove to be an important factor in decision-making. SYNTAX-based tools (residual and post-CABG SYNTAX Scores), as measures of completeness of revascularisation, show significant promise, and may aid the heart team in determining a reasonable level of revascularisation. The ongoing development of a non-invasive functional SYNTAX Score would undoubtedly prove to be a significant development in streamlining and simplifying the heart team process.

References

Footnotes

  • Contributors VF and PWS critically reviewed the literature and cowrote the manuscript. SH and APK critically reviewed the manuscript.

  • Provenance and peer review Commissioned; externally peer reviewed.