Risk stratification and risk models in revascularisation

Summary

Health care systems have been developed with several goals in mind. Such goals include among others development of strategies for prevention of disease such as the provision of clean air, water, and adequate nutrition. Similar goals would include prevention of disease by emphasizing modification of identifiable risk factors in populations of patients such as implementation of no smoking guidelines and regulations or the development of “poly-pills’, the latter of which have been used in several country wide initiatives. A different non population based goal is centered on more personalized medicine whereby the universe of a single patient can be studied and strategies for improved outcome implemented. Quality of care metrics can be formulated for each of these goals. When developing personalized strategies, the results of randomized controlled studies are considered optimal. It is important to remember that the results of those trials related to and identify an average treatment effect which may not be applicable for the individual patient to be treated. To develop a personalized patient centric approach, data from the large randomized clinical trials can be used to test of validity and then identify individual predictive models with a good fit to achieve the desired goal.

These issues are an integral part of discussions and approaches to the field of coronary artery disease by virtue of the fact that this involves a group of patients who contribute to the majority of cardiovascular deaths. The issues are further complicated that there are multiple considerations including preventive optimal medical care, and coronary revascularization by different and either competing or collaborative approaches namely surgical versus percutaneous revascularization in addition to optimal medical therapy. As the number of therapeutic strategies continues to increase, both pharmaceutical and device, it is increasingly important to as closely as possible optimize care for the specific individual patient. In this regard, it also must be remembered that there are additional considerations of resource utilization and cost to be considered. For the individual patient the process of optimizing care when multiple potential strategies are possible, involves frank and open discussion about the risks and benefits of each. This concept has become part of the lexicon in modern cardiovascular disease known as “ shared decision making” in which the emphasis is on the discussion carried out in ways that are open and at a level of understanding on the part of the patients and families. A variety of technologies have become popularized to facilitate this discussion. [11. O’Neill ES, Grande SW, Sherman A, Elwyn G, Coylewright M. Availability of patient decision aids for stroke prevention in atrial fibrillation: A systematic review. Am Heart J. 2017;191:1-11. ] Some of the stratification models will be useful in deciding on specific therapeutic strategies to be recommended; for example there may be some patients in whom PCI is not recommended based upon risk in which case surgical revascularization alone will be indicated. At the heart of this strategy is the concept of the Heart Team which has now become embedded in Professional Societal Guidelines [22. Holmes DR, Jr, Mack MJ. Rational Dispersion for the Introduction of Transcatheter Mitral Valve Repair Into Clinical Practice: Lessons Learned From TAVR. Jama. 2019;321(11):1043-1044. ] The risk stratification models may also be used to identify post procedural regimens that will be recommended, for example the duration of dual antiplatelet therapy. [33. Nammas W, Kiviniemi T, Schlitt A, Rubboli A, Valencia J, Lip GYH, Karjalainen PP, Biancari F, Juhani Airaksinen KE. Value of DAPT score to predict adverse outcome in patients with atrial fibrillation undergoing percutaneous coronary intervention: A post-hoc analysis from the AFCAS registry. Int J Cardiol. 2018;253:35-39. ] For each of these scenarios, risk stratification is essential and is the focus of this chapter.

Introduction

Risk stratification is an important aspect of clinical medicine, being central to every decision made by a practicing physician particularly when there are several potential strategies that could be recommended or implemented. There is increasing emphasis on assessing risk in patients undergoing coronary revascularization as a consequence of several factors highlighted below.

INCREASING PATIENT COMORBIDITIES

The demographic of patients requiring revascularization is changing as a consequence of increased longevity of the general populations to be treated, a lower threshold to investigate patients presenting with symptoms suggestive of obstructive coronary artery disease (CAD), and increased technology and expertise making revascularization potentially safer and more accessible. [44. Windecker S, Neumann FJ, Jüni P, Sousa-Uva M, Falk V. Considerations for the choice between coronary artery bypass grafting and percutaneous coronary intervention as revascularization strategies in major categories of patients with stable multivessel coronary artery disease: an accompanying article of the task force of the 2018 ESC/EACTS guidelines on myocardial revascularization. Eur Heart J. 2019;40(2):204-212. , 55. Patel MR, Calhoon JH, Dehmer GJ, Grantham JA, Maddox TM, Maron DJ, Smith PK. ACC/AATS/AHA/ASE/ASNC/SCAI/SCCT/STS 2016 Appropriate Use Criteria for Coronary Revascularization in Patients With Acute Coronary Syndromes: A Report of the American College of Cardiology Appropriate Use Criteria Task Force, American Association for Thoracic Surgery, American Heart Association, American Society of Echocardiography, American Society of Nuclear Cardiology, Society for Cardiovascular Angiography and Interventions, Society of Cardiovascular Computed Tomography, and the Society of Thoracic Surgeons. J Am Coll Cardiol. 2017;69(5):570-591. ] In addition the central importance of a Heart Team approach to decision making in patients being treated for chronic CAD has been emphasized and subjected to increased scrutiny. In this discussion, patient preferences are accorded increasing value and they participate in their own individual care. For example, in patients who might be a candidate for either PCI or CABG, their decisions can reflect their individual expectations, and their individual assessment of the risks and benefits of each procedure. Another factor that while hard to quantify is also nonetheless important, relates to increasing scrutiny and attention placed on operator outcomes either surgically or by interventional colleagues. Although professional societies have published guidelines which emphasize surgical revascularization for very complex coronary anatomy such as unprotected LMCA stenosis or severe multivessel disease, not all patients in this category are deemed to be surgical candidates. Waldo et al, [66. Waldo SW, Secemsky EA, O’Brien C, Kennedy KF, Pomerantsev E, Sundt TM, 3rd, McNulty EJ, Scirica BM, Yeh RW. Surgical ineligibility and mortality among patients with unprotected left main or multivessel coronary artery disease undergoing percutaneous coronary intervention. Circulation. 2014;130(25):2295-301. ] evaluated the concept of ‘surgical ineligibility” in patients undergoing PCI in this group of patients. In 1013 patients, 22% were deemed to be “surgically ineligible”. This designation which is based on characteristics not well captured in current databases had a significant effect on outcome and was independently associated with both short and longer term mortality. As more attention is paid to optimizing individual operator outcomes, there may be a potential to assign patients to being ineligible for either surgical or percutaneous revascularization. By restricting patient selection in this way and avoiding very high risk procedures, individual operator outcomes may be improved. While difficult to quantify, this may have profound implications on patient outcomes. In addition to these changes is the increased prevalence of comorbidities such as diabetes, hypertension, and hyperlipidemia which all accelerate atherosclerosis, such that those patients presenting with CAD are now more likely to have extensive disease. Importantly, these comorbidities have also been shown to have a role in influencing clinical outcomes following revascularization; for example, age and diabetes are recognized as independent predictors of adverse outcome following percutaneous revascularization[77. Legrand VM, Garg S, Serruys PW, Virtanen KS, Szurawitzki G, Voudris V, Fontanelli A, Endersen K, Kranjec I, Rademaker T, Stefanidis CI, Wittebols K. Influence of age on the clinical outcomes of coronary revascularisation for the treatment of patients with multivessel de novo coronary artery lesions: sirolimus-eluting stent vs. coronary artery bypass surgery and bare metal stent, insight from the multicentre randomised Arterial Revascularisation Therapy Study Part I (ARTS-I) and Part II (ARTS-II). EuroIntervention. 2011;6(7):838-45. ]. It is no surprise therefore that patient comorbidities occupy prominent positions in a number of contemporary risk models. [88. Garg S, Sarno G, Garcia-Garcia HM, Girasis C, Wykrzykowska J, Dawkins KD, Serruys PW. A new tool for the risk stratification of patients with complex coronary artery disease: the Clinical SYNTAX Score. Circ Cardiovasc Interv. 2010;3(4):317-26. , 99. Ranucci M, Castelvecchio S, Menicanti L, Frigiola A, Pelissero G. Risk of assessing mortality risk in elective cardiac operations: age, creatinine, ejection fraction, and the law of parsimony. Circulation. 2009;119(24):3053-61. , 1010. Peterson ED, Dai D, DeLong ER, Brennan JM, Singh M, Rao SV, Shaw RE, Roe MT, Ho KK, Klein LW, Krone RJ, Weintraub WS, Brindis RG, Rumsfeld JS, Spertus JA. Contemporary mortality risk prediction for percutaneous coronary intervention: results from 588,398 procedures in the National Cardiovascular Data Registry. J Am Coll Cardiol. 2010;55(18):1923-32. , 1111. Chen SL, Chen JP, Mintz G, Xu B, Kan J, Ye F, Zhang J, Sun X, Xu Y, Jiang Q, Zhang A, Stone GW. Comparison between the NERS (New Risk Stratification) score and the SYNTAX (Synergy between Percutaneous Coronary Intervention with Taxus and Cardiac Surgery) score in outcome prediction for unprotected left main stenting. JACC Cardiovasc Interv. 2010;3(6):632-41. , 1212. Capodanno D, Miano M, Cincotta G, Caggegi A, Ruperto C, Bucalo R, Sanfilippo A, Capranzano P, Tamburino C. EuroSCORE refines the predictive ability of SYNTAX score in patients undergoing left main percutaneous coronary intervention. Am Heart J. 2010;159(1):103-9. , 1313. Farooq V, van Klaveren D, Steyerberg EW, Meliga E, Vergouwe Y, Chieffo A, Kappetein AP, Colombo A, Holmes DR, Mack M, Feldman T, Morice MC, Stahle E, Onuma Y, Morel MA, Garcia-Garcia HM, van Es GA, Dawkins KD, Mohr FW, Serruys PW. Anatomical and clinical characteristics to guide decision making between coronary artery bypass surgery and percutaneous coronary intervention for individual patients: development and validation of SYNTAX score II. Lancet. 2013;381(9867):639-650. ]

TECHNOLOGICAL ADVANCES

The routine use of drug-eluting stents has dramatically effected revascularization by virtue of their now well documented improvement in need for repeat revascularization, the reduction in target vessel failure and in some patient groups improvement in survival compared with bare metal stents. These improvements subsequently prompted an expansion in the indications for PCI, such that bifurcation lesions, chronic total occlusions, left main coronary artery disease and multivessel disease are now increasingly treated with PCI, whereas previously these lesion subsets were deemed more appropriate for surgical revascularization. There are still lesion subsets which provide clinical and technical challenges, such as severe lesion calcifications, lesions with very high thrombus burden, patients with severe hemodynamic instability or severe heart failure and those with recurrent diffuse in-stent restenosis. Technical improvements with wider spread use of rotational and orbital atherectomy, laser assisted procedures, lithotripsy balloons, and drug eluting balloons have all been applied in these adverse patient and lesion subsets. From the circulatory support standpoint, there is increasing availability and application of several options ranging from Impella to maximal circulatory support with percutaneous bypass. Improvements in diagnostic strategies, with the use of IVUS, OCT and IFR/FFR have helped select patients and lesions which would benefit from revascularization as well as allow assessment of the adequacy of the initial procedure. Ultimately all these advances have ensured that, technically, PCI can be used to treat a sizeable majority of coronary lesions. It must be emphasized that the feasibility of a procedure is not necessarily the best indication for its performance. The combination of improved technology plus the aforementioned maxim makes it imperative that patient selection and expectations regarding risk and benefits be central to decision making.

GUIDELINE RECOMMENDATIONS

Further reading: A compendium of international guidelines for interventional practice

Recent guideline recommendations from regulatory authorities in Europe and the US have both formally recognized the importance of risk stratifying patients with coronary artery disease appropriately. The joint European Society of Cardiology and European Association of Cardiothoracic Surgeons (ESC/EACTS) guidelines for myocardial revascularization, and the 2017 guidelines for PCI from the American College of Cardiology (ACC)/ American Heart Association (AHA)/SCAI advocated specific risk models, and make reference to these models throughout their subsequent recommendations for the management of stable CAD [55. Patel MR, Calhoon JH, Dehmer GJ, Grantham JA, Maddox TM, Maron DJ, Smith PK. ACC/AATS/AHA/ASE/ASNC/SCAI/SCCT/STS 2016 Appropriate Use Criteria for Coronary Revascularization in Patients With Acute Coronary Syndromes: A Report of the American College of Cardiology Appropriate Use Criteria Task Force, American Association for Thoracic Surgery, American Heart Association, American Society of Echocardiography, American Society of Nuclear Cardiology, Society for Cardiovascular Angiography and Interventions, Society of Cardiovascular Computed Tomography, and the Society of Thoracic Surgeons. J Am Coll Cardiol. 2017;69(5):570-591. , 1414. Neumann FJ, Sousa-Uva M, Ahlsson A, Alfonso F, Banning AP, Benedetto U, Byrne RA, Collet JP, Falk V, Head SJ, Juni P, Kastrati A, Koller A, Kristensen SD, Niebauer J, Richter DJ, Seferovic PM, Sibbing D, Stefanini GG, Windecker S, Yadav R, Zembala MO, Group ESCSD. 2018 ESC/EACTS Guidelines on myocardial revascularization. Eur Heart J. 2019;40(2):87-165. ]. These guidelines have also emphasized as a Class I indication implementation of a Heart Team approach in patients with more complex and extensive CAD, particularly patients with diabetes mellitus and those with decreased LV function and extensive widespread ischaemia.

Benefits of risk stratification

Overall risk stratification provides several benefits for both patients and clinicians.

In the short-term, they can provide supplementary information to clinicians which can help guide treatment strategy [1515. Capodanno D, Capranzano P, Di Salvo ME, Caggegi A, Tomasello D, Cincotta G, Miano M, Patané M, Tamburino C, Tolaro S, Patané L, Calafiore AM, Tamburino C. Usefulness of SYNTAX score to select patients with left main coronary artery disease to be treated with coronary artery bypass graft. JACC Cardiovasc Interv. 2009;2(8):731-8. , 1616. Serruys PW, Morice MC, Kappetein AP, Colombo A, Holmes DR, Mack MJ, Ståhle E, Feldman TE, van den Brand M, Bass EJ, Van Dyck N, Leadley K, Dawkins KD, Mohr FW. Percutaneous coronary intervention versus coronary-artery bypass grafting for severe coronary artery disease. N Engl J Med. 2009;360(10):961-72. ]. In addition, this stratification of risk can assist in ensuring that patients are more adequately informed about the risks/benefits of the revascularization strategies available, allowing them to make more informed decisions [1717. Federspiel JJ, Stearns SC, van Domburg RT, Sheridan BC, Lund JL, Serruys PW. Risk-benefit trade-offs in revascularisation choices. EuroIntervention. 2011;6(8):936-41. ]. It must be remembered that patient centric decision making is always the goal and that ultimately it is the clinician’s duty to convey full and understandable information to their patients and their families or caregivers [1818. Jørgensen KJ, Brodersen J, Hartling OJ, Nielsen M, Gøtzsche PC. Informed choice requires information about both benefits and harms. J Med Ethics. 2009;35(4):268-9. ].

Over the longer-term, risk models can be used to measure the success of treatments, while also helping to identify future directions to further improve clinical outcomes such as the need for continued more aggressive dual antiplatelet therapy [1919. Mauri L, Kereiakes DJ, Yeh RW, Driscoll-Shempp P, Cutlip DE, Steg PG, Normand SL, Braunwald E, Wiviott SD, Cohen DJ, Holmes DR, Jr, Krucoff MW, Hermiller J, Dauerman HL, Simon DI, Kandzari DE, Garratt KN, Lee DP, Pow TK, Ver Lee P, Rinaldi MJ, Massaro JM. Twelve or 30 months of dual antiplatelet therapy after drug-eluting stents. N Engl J Med. 2014;371(23):2155-66. ] . Clinical governance and the increasing requirement to report clinical performance (and complications) publicly have also propelled the need to risk stratify patients thereby allowing a useful comparison of performance to be made between clinicians (and institutions) and the standards dictated by regulatory authorities [2020. Califf RM, Peterson ED, Gibbons RJ, Garson A, Jr, Brindis RG, Beller GA, Smith SC, Jr Integrating quality into the cycle of therapeutic development. J Am Coll Cardiol. 2002;40(11):1895-901. ] . Finally their importance has been enhanced as it becomes increasingly essential for clinicians to be able to justify clinical decisions to patients, peers and regulatory bodies.

Types of risk assessment

Risk assessment can be largely divided into two main categories:

QUALITATIVE RISK STRATIFICATION

Qualitative risk stratification is the simplest form of risk assessment, and is sometime referred to as "the end of the bed test." This subjective assessment relies on a large number of poorly defined features which have been learned over years of clinical experience. It may be part of the assessment previously discussed that places a patient in the category of 'surgically ineligible' [66. Waldo SW, Secemsky EA, O’Brien C, Kennedy KF, Pomerantsev E, Sundt TM, 3rd, McNulty EJ, Scirica BM, Yeh RW. Surgical ineligibility and mortality among patients with unprotected left main or multivessel coronary artery disease undergoing percutaneous coronary intervention. Circulation. 2014;130(25):2295-301. ]. Similarly some patients may be deemed to be not candidates for percutaneous intervention by virtue of factors in this poorly defined "end of the bed test". From a clinician’s perspective this form of risk stratification can be performed instantaneously and sub-consciously without reliance on external aids. From a patient’s perspective this assessment has the greatest sensitivity as all factors relevant to the estimation of risk in that particular individual can be considered, rather than the select list of variables incorporated within a risk model. Of note, this subjective qualitative assessment also allows risk to be calculated and tailored to the expertise and experience of the operator in patients similar to the one currently being evaluated by the physician performing the procedure. Disadvantages of this approach to risk assessment are its dependence on either the operator’s prior experience or the perception of the operator on the outcome of similar patients treated by him or her, and its high intra- and inter-observer variability. Other disadvantages are the potential use of this approach to “game the system” and improve operator outcomes by restricting patient selection criteria. In addition, this approach while valuable is very difficult to teach fellows in training or physicians who have a different level of experience.

QUANTITATIVE RISK STRATIFICATION

Quantitative risk stratification can be performed using a variety of risk models which incorporate clinical variables which have been derived from large historical patient databases [99. Ranucci M, Castelvecchio S, Menicanti L, Frigiola A, Pelissero G. Risk of assessing mortality risk in elective cardiac operations: age, creatinine, ejection fraction, and the law of parsimony. Circulation. 2009;119(24):3053-61. , 1010. Peterson ED, Dai D, DeLong ER, Brennan JM, Singh M, Rao SV, Shaw RE, Roe MT, Ho KK, Klein LW, Krone RJ, Weintraub WS, Brindis RG, Rumsfeld JS, Spertus JA. Contemporary mortality risk prediction for percutaneous coronary intervention: results from 588,398 procedures in the National Cardiovascular Data Registry. J Am Coll Cardiol. 2010;55(18):1923-32. , 2121. Roques F, Nashef SA, Michel P, Gauducheau E, de Vincentiis C, Baudet E, Cortina J, David M, Faichney A, Gabrielle F, Gams E, Harjula A, Jones MT, Pintor PP, Salamon R, Thulin L. Risk factors and outcome in European cardiac surgery: analysis of the EuroSCORE multinational database of 19030 patients. Eur J Cardiothorac Surg 1999;15(6):816-22; discussion 822-3. , 2222. Singh M, Peterson ED, Milford-Beland S, Rumsfeld JS, Spertus JA. Validation of the Mayo clinic risk score for in-hospital mortality after percutaneous coronary interventions using the national cardiovascular data registry. Circ Cardiovasc Interv. 2008;1(1):36-44. ]. The majority of these models rely on clinical variables whose presence is recorded through binary responses, resulting in a high level of reproducibility. By contrast those models such as the ACC/AHA lesion score [2323. Ryan TJ, Bauman WB, Kennedy JW, Kereiakes DJ, King SB, 3rd, McCallister BD, Smith SC, Jr, Ullyot DJ. Guidelines for percutaneous transluminal coronary angioplasty. A report of the American Heart Association/American College of Cardiology Task Force on Assessment of Diagnostic and Therapeutic Cardiovascular Procedures (Committee on Percutaneous Transluminal Coronary Angioplasty). Circulation. 1993;88(6):2987-3007. ] and the SYNTAX score [2424. Sianos G, Morel MA, Kappetein AP, Morice MC, Colombo A, Dawkins K, van den Brand M, Van Dyck N, Russell ME, Mohr FW, Serruys PW. The SYNTAX Score: an angiographic tool grading the complexity of coronary artery disease. EuroIntervention. 2005;1(2):219-27. ] (available at https://syntaxscore2020.com/) which include angiographic variables continue to have documented intra- and inter-observer variability [2525. Serruys PW, Onuma Y, Garg S, Sarno G, van den Brand M, Kappetein AP, Van Dyck N, Mack M, Holmes D, Feldman T, Morice MC, Colombo A, Bass E, Leadley K, Dawkins KD, van Es GA, Morel MA, Mohr FW. Assessment of the SYNTAX score in the Syntax study. EuroIntervention. 2009;5(1):50-6. , 2626. Garg S, Girasis C, Sarno G, Goedhart D, Morel MA, Garcia-Garcia HM, Bressers M, van Es GA, Serruys PW. The SYNTAX score revisited: a reassessment of the SYNTAX score reproducibility. Catheter Cardiovasc Interv. 2010;75(6):946-52. ]. The angiographic SYNTAX Score has been improved by the addition of clinical variables which improves the ability to risk stratify patients being evaluated for revascularization. These quantitative risk models have three other notable disadvantages, which need to be considered when interpreting their results:

1. Unlike qualitative risk assessment which while very subjective has the potential to have better sensitivity for the individual patient in question, the limited numbers of variables in quantitative models results in a lack of sensitivity to accurately predict risk in that individual patient. Rather these models are more suited to predicting risk for a population of patients with similar comorbidities.
2. Quantitative models are limited by a practical need to balance the number of variables involved to allow a meaningful estimation of risk, while on the other hand ensuring the model is not too complicated or time consuming to use which would reduce user uptake. Faced with busy increasingly demanding schedules, models must be able to be used at the point of care to have maximal value. Minimizing the number of variables also reduces the chances of co-linearity between independent variables, which can result in redundant information being collected [99. Ranucci M, Castelvecchio S, Menicanti L, Frigiola A, Pelissero G. Risk of assessing mortality risk in elective cardiac operations: age, creatinine, ejection fraction, and the law of parsimony. Circulation. 2009;119(24):3053-61. ] , while also increasing the chances of "over-fitting" the model, and thereby reducing the overall accuracy of the results.
3. These quantitative risk models rely on large patient databases to derive the variables needed to estimate risk, however the very nature of this means that risk estimates are based on historical retrospective data, which may no longer be so relevant in contemporary practice. This is particularly important in rapidly evolving fields. For example The European System for Cardiac Operative Risk Evaluation (EuroSCORE) was developed using data collected prior to 2000. Technological and pharmacological developments in surgical practice since that time have improved surgical outcomes and reduced perioperative mortality, such that it is no surprise that repeated evaluations of the EuroSCORE indicate that it over-estimates risk by a factor of 2 or 3 [2727. Bhatti F, Grayson AD, Grotte G, Fabri BM, Au J, Jones M, Bridgewater B. The logistic EuroSCORE in cardiac surgery: how well does it predict operative risk. Heart. 2006;92(12):1817-20. ]. In contrast to this, The Society of Thoracic Surgeons (STS) score is also derived from a large patient database, however unlike the EuroSCORE, the STS calculator is being periodically recalibrated to ensure its results remain relevant to contemporary practice [2828. Shahian DM, O’Brien SM, Filardo G, Ferraris VA, Haan CK, Rich JB, Normand SL, DeLong ER, Shewan CM, Dokholyan RS, Peterson ED, Edwards FH, Anderson RP. The Society of Thoracic Surgeons 2008 cardiac surgery risk models: part 1--coronary artery bypass grafting surgery. Ann Thorac Surg. 2009;88(1 Suppl):S2-22. ]. It must be remembered that the specific risk score must be matched with the specific indication for the procedure. For example, the STS risk score for valvular heart disease was developed for patients considered for surgical aortic valve replacement; it is being used in the evaluation of patient for transcatheter aortic valve replacement [2929. Dewey TM, Brown D, Ryan WH, Herbert MA, Prince SL, Mack MJ. Reliability of risk algorithms in predicting early and late operative outcomes in high-risk patients undergoing aortic valve replacement. J Thorac Cardiovasc Surg. 2008;135(1):180-7. ]. For these latter patients, the surgical score may need to be revised.

Contemporary risk for elective revascularisation

Since the mid-1990s numerous different risk models have been proposed for patients undergoing revascularization; however, apart from the EuroSCORE most of these initial risk scores have now been superseded by more contemporary models. With respect to the current array of risk models, these can be grouped according to:

1. The type of information used to quantify risk. This can range from data (used either in isolation or combination) related to clinical variables such as urgency of presentation and comorbidities; complexity of coronary anatomy, and/or procedural factors.
2. The modality of revascularization for which risk can be estimated. Some risk models have been validated in patients undergoing only one form of revascularization, whilst others have been assessed in patients undergoing either PCI or coronary artery bypass grafting (CABG). It follows that those models which have been assessed in patients undergoing any form of revascularization can be calculated prior to the selection of treatment strategy. These scores evaluate long-term risk which is important for patient selection.
3. Models have also been developed to estimate the effect of specific revascularization strategies on subsequent outcome. The recent Residual SYNTAX Score was developed with that in mind [3030. Farooq V, Serruys PW, Bourantas CV, Zhang Y, Muramatsu T, Feldman T, Holmes DR, Mack M, Morice MC, Ståhle E, Colombo A, de Vries T, Morel MA, Dawkins KD, Kappetein AP, Mohr FW. Quantification of incomplete revascularization and its association with five-year mortality in the synergy between percutaneous coronary intervention with taxus and cardiac surgery (SYNTAX) trial validation of the residual SYNTAX score. Circulation. 2013;128(2):141-51. ] . This score allows assessment of the completeness of revascularization on subsequent outcome after either PCI or CABG for the treatment of severe complex multivessel CAD. In patients treated with PCI, complete revascularization provides the best longer term outcome of freedom from MACCE; in contrast, increasing degrees of incomplete revascularization with increasing residual untreated disease results in a gradation in worsening outcome.

Current Scoring Systems

As mentioned the scoring systems include a variety of variables, and are used to identify the risk/benefit ratio of specific treatment strategies, in this case guideline based medical therapy alone, PCI or CABG. They are also focused on variable outcomes, for example prediction of outcomes such as mortality, MACCE, or quality of life issues. It must be remembered that while these scores can be used to describe and predict outcomes in large groups of patients, for the individual patient they are limited; the patient either experiences the predicted outcome or does not in a dichotomous manner. Be that as it may, the risk scores are helpful in counseling patients and their families, in developing professional guidelines and will be used for comparative effectiveness research which will guide some aspects of reimbursement in the future. Certainly, these scoring systems should be used by the Heart Team in formulating recommendations for the specific patient.

Each score has advantages and disadvantages. Those which use only demographic clinical pre procedural variables are the easiest to use, but have the disadvantage of not including all of the important information such as angiographic findings.

Clinical based scores only assess clinical variables, and therefore do not require a coronary angiogram to facilitate the quantification of risk. This offers the ability to calculate risk relatively quickly at the bedside, and advantageously these models usually include variables which are not subject to user interpretation, thereby ensuring excellent reproducibility.

EuroSCORE

The additive EuroSCORE [3131. Nashef SA, Roques F, Michel P, Gauducheau E, Lemeshow S, Salamon R. European system for cardiac operative risk evaluation (EuroSCORE). Eur J Cardiothorac Surg 1999;16(1):9-13. ] which quantifies risk using 17 clinical variables ( Table 1 ), was initially developed to predict in-hospital, and long-term mortality in patients undergoing cardiac surgery [3131. Nashef SA, Roques F, Michel P, Gauducheau E, Lemeshow S, Salamon R. European system for cardiac operative risk evaluation (EuroSCORE). Eur J Cardiothorac Surg 1999;16(1):9-13. , 3232. Toumpoulis IK, Anagnostopoulos CE, DeRose JJ, Swistel DG. European system for cardiac operative risk evaluation predicts long-term survival in patients with coronary artery bypass grafting. Eur J Cardiothorac Surg. 2004;25(1):51-8. , 3333. De Maria R, Mazzoni M, Parolini M, Gregori D, Bortone F, Arena V, Parodi O. Predictive value of EuroSCORE on long term outcome in cardiac surgery patients: a single institution study. Heart 2005;91(6):779-84. ]. Validation studies documented that this additive model underestimated risk in those at highest risk. Accordingly it has been replaced. This group of scores continues to evolve. The EuroSCORE II is an update of the logistic EuroScore and used data from more contemporary surgical practice, which uses the same clinical variables to quantify risk, however the calculation is more complicated necessitating the use of an on-line calculator (available at http://www.euroscore.org/calc.html) [3434. Roques F, Michel P, Goldstone AR, Nashef SA. The logistic EuroSCORE. Eur Heart J. 2003;24(9):881-2. ]. This limits the widespread application of this score at the point of care. Further to its assessment and validation in patients undergoing surgical revascularisation, the EuroSCORE has also been evaluated in numerous studies of patients undergoing PCI [1616. Serruys PW, Morice MC, Kappetein AP, Colombo A, Holmes DR, Mack MJ, Ståhle E, Feldman TE, van den Brand M, Bass EJ, Van Dyck N, Leadley K, Dawkins KD, Mohr FW. Percutaneous coronary intervention versus coronary-artery bypass grafting for severe coronary artery disease. N Engl J Med. 2009;360(10):961-72. , 3535. Serruys PW, Onuma Y, Garg S, Vranckx P, De Bruyne B, Morice MC, Colombo A, Macaya C, Richardt G, Fajadet J, Hamm C, Schuijer M, Rademaker T, Wittebols K, Stoll HP. 5-year clinical outcomes of the ARTS II (Arterial Revascularization Therapies Study II) of the sirolimus-eluting stent in the treatment of patients with multivessel de novo coronary artery lesions. J Am Coll Cardiol. 2010;55(11):1093-101. , 3636. Witberg, G., et al., The Prognostic Effects of Coronary Disease Severity and Completeness of Revascularization on Mortality in Patients Undergoing Transcatheter Aortic Valve Replacement. JACC Cardiovasc Interv, 2017. 10(14): p. 1428-1435. , 3737. Onuma Y, Girasis C, Piazza N, Garcia-Garcia HM, Kukreja N, Garg S, Eindhoven J, Cheng JM, Valgimigli M, van Domburg R, Serruys PW. Long-term clinical results following stenting of the left main stem: insights from RESEARCH (Rapamycin-Eluting Stent Evaluated at Rotterdam Cardiology Hospital) and T-SEARCH (Taxus-Stent Evaluated at Rotterdam Cardiology Hospital) Registries. JACC Cardiovasc Interv. 2010;3(6):584-94. , 3838. Rodés-Cabau J, Deblois J, Bertrand OF, Mohammadi S, Courtis J, Larose E, Dagenais F, Déry JP, Mathieu P, Rousseau M, Barbeau G, Baillot R, Gleeton O, Perron J, Nguyen CM, Roy L, Doyle D, De Larochellière R, Bogaty P, Voisine P. Nonrandomized comparison of coronary artery bypass surgery and percutaneous coronary intervention for the treatment of unprotected left main coronary artery disease in octogenarians. Circulation. 2008;118(23):2374-81. ], the majority of which have specifically enrolled patients with LM disease. Nevertheless, irrespective of disease severity, the EuroSCORE (additive or logistic) has consistently been identified as an independent predictor of mortality and/or major adverse cardiovascular events (MACCE) at follow-up of up to 5-years [1616. Serruys PW, Morice MC, Kappetein AP, Colombo A, Holmes DR, Mack MJ, Ståhle E, Feldman TE, van den Brand M, Bass EJ, Van Dyck N, Leadley K, Dawkins KD, Mohr FW. Percutaneous coronary intervention versus coronary-artery bypass grafting for severe coronary artery disease. N Engl J Med. 2009;360(10):961-72. , 3535. Serruys PW, Onuma Y, Garg S, Vranckx P, De Bruyne B, Morice MC, Colombo A, Macaya C, Richardt G, Fajadet J, Hamm C, Schuijer M, Rademaker T, Wittebols K, Stoll HP. 5-year clinical outcomes of the ARTS II (Arterial Revascularization Therapies Study II) of the sirolimus-eluting stent in the treatment of patients with multivessel de novo coronary artery lesions. J Am Coll Cardiol. 2010;55(11):1093-101. , 3636. Witberg, G., et al., The Prognostic Effects of Coronary Disease Severity and Completeness of Revascularization on Mortality in Patients Undergoing Transcatheter Aortic Valve Replacement. JACC Cardiovasc Interv, 2017. 10(14): p. 1428-1435. , 3737. Onuma Y, Girasis C, Piazza N, Garcia-Garcia HM, Kukreja N, Garg S, Eindhoven J, Cheng JM, Valgimigli M, van Domburg R, Serruys PW. Long-term clinical results following stenting of the left main stem: insights from RESEARCH (Rapamycin-Eluting Stent Evaluated at Rotterdam Cardiology Hospital) and T-SEARCH (Taxus-Stent Evaluated at Rotterdam Cardiology Hospital) Registries. JACC Cardiovasc Interv. 2010;3(6):584-94. , 3838. Rodés-Cabau J, Deblois J, Bertrand OF, Mohammadi S, Courtis J, Larose E, Dagenais F, Déry JP, Mathieu P, Rousseau M, Barbeau G, Baillot R, Gleeton O, Perron J, Nguyen CM, Roy L, Doyle D, De Larochellière R, Bogaty P, Voisine P. Nonrandomized comparison of coronary artery bypass surgery and percutaneous coronary intervention for the treatment of unprotected left main coronary artery disease in octogenarians. Circulation. 2008;118(23):2374-81. ]. Notably, in those studies which also included a surgical control group, such as the SYNTAX study, and the MAIN-COMPARE study the EuroSCORE has also been shown to an independent predictor of major adverse cardiovascular and cerebrovascular events (MACCE) in surgical patients [3636. Morice MC, Serruys PW, Kappetein AP, Feldman TE, Ståhle E, Colombo A, Mack MJ, Holmes DR, Torracca L, van Es GA, Leadley K, Dawkins KD, Mohr F. Outcomes in patients with de novo left main disease treated with either percutaneous coronary intervention using paclitaxel-eluting stents or coronary artery bypass graft treatment in the Synergy Between Percutaneous Coronary Intervention with TAXUS and Cardiac Surgery (SYNTAX) trial. Circulation. 2010;121(24):2645-53. , 3838. Rodés-Cabau J, Deblois J, Bertrand OF, Mohammadi S, Courtis J, Larose E, Dagenais F, Déry JP, Mathieu P, Rousseau M, Barbeau G, Baillot R, Gleeton O, Perron J, Nguyen CM, Roy L, Doyle D, De Larochellière R, Bogaty P, Voisine P. Nonrandomized comparison of coronary artery bypass surgery and percutaneous coronary intervention for the treatment of unprotected left main coronary artery disease in octogenarians. Circulation. 2008;118(23):2374-81. ].

The only study which has assessed outcomes according to the EuroSCORE in patients randomised to treatment with PCI or CABG is the SYNTAX study. At 1-year follow-up, the additive EuroSCORE was shown to be an independent predictor of MACCE in patients treated with PCI (OR: 1.12; 95% CI [1.00-1.25], p=0.0045) and CABG (OR: 1.19; 95% CI [1.04-1.35], p=0.009) [3636. Morice MC, Serruys PW, Kappetein AP, Feldman TE, Ståhle E, Colombo A, Mack MJ, Holmes DR, Torracca L, van Es GA, Leadley K, Dawkins KD, Mohr F. Outcomes in patients with de novo left main disease treated with either percutaneous coronary intervention using paclitaxel-eluting stents or coronary artery bypass graft treatment in the Synergy Between Percutaneous Coronary Intervention with TAXUS and Cardiac Surgery (SYNTAX) trial. Circulation. 2010;121(24):2645-53. ]. Likewise, registry data from Rodés-Cabau et al identified a EuroSCORE ≥9 to be the best predictor of MACCE at 23-months follow-up after PCI or CABG amongst 249 octogenarians with LM disease [3838. Rodés-Cabau J, Deblois J, Bertrand OF, Mohammadi S, Courtis J, Larose E, Dagenais F, Déry JP, Mathieu P, Rousseau M, Barbeau G, Baillot R, Gleeton O, Perron J, Nguyen CM, Roy L, Doyle D, De Larochellière R, Bogaty P, Voisine P. Nonrandomized comparison of coronary artery bypass surgery and percutaneous coronary intervention for the treatment of unprotected left main coronary artery disease in octogenarians. Circulation. 2008;118(23):2374-81. ] . The much larger MAIN-COMPARE registry also reported similar findings amongst over 1500 patients with LM disease who were follow-up for a median of 3.1 years; in this study a EuroSCORE ≥6 was shown to be an independent predictor of mortality following either PCI or CABG [3939. Min SY, Park DW, Yun SC, Kim YH, Lee JY, Kang SJ, Lee SW, Lee CW, Kim JJ, Park SW, Park SJ. Major predictors of long-term clinical outcomes after coronary revascularization in patients with unprotected left main coronary disease: analysis from the MAIN-COMPARE study. Circ Cardiovasc Interv. 2010;3(2):127-33. ].

The ability of the EuroSCORE to identify patients at high-risk of adverse events following PCI has also been demonstrated in those with coronary disease not involving the LM stem. Serruys et al demonstrated that logistic EuroSCORE was an independent predictor of MACCE out to 5-years follow-up amongst patients with 2- or 3- vessel disease who were enrolled in the ARTS-II study; in this study, it must be remembered that LM lesions were excluded [3535. Serruys PW, Onuma Y, Garg S, Vranckx P, De Bruyne B, Morice MC, Colombo A, Macaya C, Richardt G, Fajadet J, Hamm C, Schuijer M, Rademaker T, Wittebols K, Stoll HP. 5-year clinical outcomes of the ARTS II (Arterial Revascularization Therapies Study II) of the sirolimus-eluting stent in the treatment of patients with multivessel de novo coronary artery lesions. J Am Coll Cardiol. 2010;55(11):1093-101. ]. In addition, Romagnoli et al [4040. Romagnoli E, Burzotta F, Trani C, Siviglia M, Biondi-Zoccai GG, Niccoli G, Leone AM, Porto I, Mazzari MA, Mongiardo R, Rebuzzi AG, Schiavoni G, Crea F. EuroSCORE as predictor of in-hospital mortality after percutaneous coronary intervention. Heart. 2009;95(1):43-8. ] reported that the EuroSCORE was an independent predictor of in-hospital mortality amongst over 1,100 patients, 70% of whom had single vessel disease. The C-statistic for the prediction of in-hospital mortality using the EuroSCORE in this population was 0.91 [4040. Romagnoli E, Burzotta F, Trani C, Siviglia M, Biondi-Zoccai GG, Niccoli G, Leone AM, Porto I, Mazzari MA, Mongiardo R, Rebuzzi AG, Schiavoni G, Crea F. EuroSCORE as predictor of in-hospital mortality after percutaneous coronary intervention. Heart. 2009;95(1):43-8. ].

In summary, while acknowledging that most of the data have been derived from non-randomised observational studies in patients with LM disease, the findings do suggest that the EuroSCORE is a valuable tool in the individual assessment of risk prior to the selection of revascularization strategy. Furthermore, as patients with a high EuroSCORE are at risk of adverse events following PCI or CABG, the EuroSCORE has a limited role in helping to select a treatment strategy.

Society of Thoracic Surgeons (STS) Score [2929. Dewey TM, Brown D, Ryan WH, Herbert MA, Prince SL, Mack MJ. Reliability of risk algorithms in predicting early and late operative outcomes in high-risk patients undergoing aortic valve replacement. J Thorac Cardiovasc Surg. 2008;135(1):180-7. ]

This score is widely used and has been validated in predicting both in-hospital or 30 day mortality and morbidity in patients undergoing surgical coronary revascularization and combined surgical coronary revascularization and valvular heart surgery. It has been widely applied in the field of transcatheter aortic valve replacement. It forms of the basis of selecting patients for enrollment in current trials TAVI. Initial experience with TAVI was gained in patients defined as either high risk or inoperable based upon the STS score. Current trials are enrolling intermediate risk patients again based in large part on an arbitrary STS score such as 8. It must remembered in this regard, that the STS score used in this way is being applied in a group of patients who are being considered for a catheter based strategy whereas, the score was based on surgical patient groups. Currently efforts are underway to develop a specific equivalent score for TAVI patients.

Age, Creatinine, Ejection Fraction (ACEF) score

The ACEF score predicts risk by combining three clinical variables (patient age, ejection fraction [1010. Peterson ED, Dai D, DeLong ER, Brennan JM, Singh M, Rao SV, Shaw RE, Roe MT, Ho KK, Klein LW, Krone RJ, Weintraub WS, Brindis RG, Rumsfeld JS, Spertus JA. Contemporary mortality risk prediction for percutaneous coronary intervention: results from 588,398 procedures in the National Cardiovascular Data Registry. J Am Coll Cardiol. 2010;55(18):1923-32. ] and serum creatinine) in the formula [patient age/ejection fraction (%)] + [1 if creatinine >2 mg/dl]. The score was initially developed for patients undergoing elective CABG, and was validated in a single-centre study which included an initial data set of 4,557 patients, and a subsequent validation series of 4,091 patients [99. Ranucci M, Castelvecchio S, Menicanti L, Frigiola A, Pelissero G. Risk of assessing mortality risk in elective cardiac operations: age, creatinine, ejection fraction, and the law of parsimony. Circulation. 2009;119(24):3053-61. ] . Results demonstrated a similar accuracy and calibration for the prediction of in-hospital mortality with the ACEF score when compared with other more established and complicated surgical risk scores such as the EuroSCORE and the Cleveland Clinic Score. A second larger study which assessed outcomes in approximately 30,000 patients from 15 different centres confirmed the superior predictive ability of the ACEF score when compared with the additive and logistic EuroSCORE, with results consistent in patients at low, intermediate and high-risk [4141. Ranucci M, Castelvecchio S, Conte M, Megliola G, Speziale G, Fiore F, Guarracino F, Scolletta S, Escobar RM, Falco M, Bignami E, Landoni G. The easier, the better: age, creatinine, ejection fraction score for operative mortality risk stratification in a series of 29,659 patients undergoing elective cardiac surgery. J Thorac Cardiovasc Surg. 2011;142(3):581-6. ].

The ability of the ACEF score to stratify risk can also be extended to patients having PCI. Wykrzykowska et al assessed the impact of the ACEF score on outcomes amongst 1,208 patients undergoing percutaneous revascularization who were enrolled in the all-comers multi-center LEADERS study. At 1-year follow-up patients in highest tertile of the ACEF score had significantly higher rates of death, cardiac death, and MI compared to those in the lower two tertiles [4242. Wykrzykowska JJ, Garg S, Onuma Y, de Vries T, Goedhart D, Morel MA, van Es GA, Buszman P, Linke A, Ischinger T, Klauss V, Corti R, Eberli F, Wijns W, Morice MC, di Mario C, van Geuns RJ, Juni P, Windecker S, Serruys PW. Value of age, creatinine, and ejection fraction (ACEF score) in assessing risk in patients undergoing percutaneous coronary interventions in the ’All-Comers’ LEADERS trial. Circ Cardiovasc Interv. 2011;4(1):47-56. ]. In addition, the ACEF score was identified as an independent predictor of MACCE and mortality. In the GLOBAL LEADERS trial, ACEF score was available in 14,941 patients, and the discrimination was helpful (C-statistic 0.72) for 2-year all-cause mortality [4343. Chichareon P, Modolo R, van Klaveren D, Takahashi K, Kogame N, Chang CC, Katagiri Y, Tomaniak M, Asano T, Spitzer E, Buszman P, Prokopczuk J, Fath-Ordoubadi F, Buysschaert I, Anderson R, Oldroyd KG, Merkely B, Garg S, Wykrzykowska JJ, Piek JJ, Juni P, Hamm C, Steg PG, Valgimigli M, Vranckx P, Windecker S, Onuma Y, Serruys PW. Predictive ability of ACEF and ACEF II score in patients undergoing percutaneous coronary intervention in the GLOBAL LEADERS study. Int J Cardiol. 2019;286:43-50. ].

Recently the ACEF Risk Score has been updated for surgical patients as the ACEF II score by including as a variable emergency surgery and pre-operative anemia. It was developed in 7011 consecutive cardiac surgical patients and then validated in 1687 from different surgical institutions. The c-statistic for the ACEF II score was 0.814 and better than the initial ACEF score. [4444. Ranucci M, Pistuddi V, Scolletta S, de Vincentiis C, Menicanti L. The ACEF II Risk Score for cardiac surgery: updated but still parsimonious. Eur Heart J. 2018;39(23):2183-2189. ]

Despite the limitations of the current data, which are retrospective in prior surgical series but now validated i, and confined to only one study in PCI patients, the ACEF and now the ACEF II scores appear to be effective and importantly parsimonious methods to risk stratify patients undergoing either percutaneous or surgical revascularization. Future prospective assessment in randomised populations treated with PCI and CABG will help define the role of these two risk scores.

FREEDOM Score

The FREEDOM score was derived from baseline characteristics and outcome of patients with diabetes mellitus and MVD without LMCAD in the FREEDOM trial, and predicts 1-year angina status and 5-year MACCE ( Table 2) [4545. Qintar M, Humphries KH, Park JE, Arnold SV, Tang Y, Jones P, Salisbury AC, Kureshi F, Farkouh ME, Fuster V, Cohen DJ, Spertus JA. Individualizing Revascularization Strategy for Diabetic Patients With Multivessel Coronary Disease. J Am Coll Cardiol. 2019;74(16):2074-2084. ]. For 1-year angina, age, BMI, presenting as ACS, history of MI, history of PVD, eGFR, LVEF, hemoglobin, and angina at baseline are used for calculating probability after PCI and CABG. Age, BMI, history of smoking, history of MI, history of stroke, requiring insulin, eGFR, and LVEF are applied for 5-year MACCE prediction. Providing risk scores after PCI and CABG can support the decision making on revascularization (PCI or CABG).

ANGIOGRAPHIC-BASED SCORES

Angiographic based risk models provide an assessment of risk which is independent of a patient’s clinical comorbidities. This form of risk stratification has inherent limitations however, as it relies on an interpretation of the coronary angiogram which is subject to intra- and inter-observer variability. [2525. Serruys PW, Onuma Y, Garg S, Sarno G, van den Brand M, Kappetein AP, Van Dyck N, Mack M, Holmes D, Feldman T, Morice MC, Colombo A, Bass E, Leadley K, Dawkins KD, van Es GA, Morel MA, Mohr FW. Assessment of the SYNTAX score in the Syntax study. EuroIntervention. 2009;5(1):50-6. , 2626. Garg S, Girasis C, Sarno G, Goedhart D, Morel MA, Garcia-Garcia HM, Bressers M, van Es GA, Serruys PW. The SYNTAX score revisited: a reassessment of the SYNTAX score reproducibility. Catheter Cardiovasc Interv. 2010;75(6):946-52. , 4646. Kleiman NS, Rodriguez AR, Raizner AE. Interobserver variability in grading of coronary arterial narrowings using the American College of Cardiology/American Heart Association grading criteria. Am J Cardiol. 1992;69(4):413-5. ]

The two angiographic risk models used in contemporary practice are:

ACC/AHA Lesion Classification

The ACC/AHA lesion classification, initially devised in 1986, and subsequently modified in 1990, uses 11 angiographic variables to categorize lesions into four groups: Type A, B1, B2 and C ( Table 3 ). Historical studies prior to the arrival of DES indicated that that ACC/AHA lesion classification did have a prognostic impact on early and late outcomes [4747. Kastrati A, Schömig A, Elezi S, Dirschinger J, Mehilli J, Schühlen H, Blasini R, Neumann FJ. Prognostic value of the modified american college of Cardiology/American heart association stenosis morphology classification for long-term angiographic and clinical outcome after coronary stent placement. Circulation. 1999;100(12):1285-90. ]. In the DES era contemporary studies, which have all been retrospective, have provided mixed results. One of the largest studies is the German Cypher registry, which enrolled over 6,700 patients with approximately 8,000 lesions, and failed to identify any definitive relationship between clinical outcomes and ACC/AHA lesion class through to 6-months follow-up [4848. Khattab AA, Hamm CW, Senges J, Toelg R, Geist V, Bonzel T, Kelm M, Levenson B, Nienaber CA, Pfannebecker T, Sabin G, Schneider S, Tebbe U, Richardt G. Prognostic value of the modified American College of Cardiology/American Heart Association lesion morphology classification for clinical outcome after sirolimus-eluting stent placement (results of the prospective multicenter German Cypher Registry). Am J Cardiol. 2008;101(4):477-82. ]. By contrast, data from studies of patients with complex disease have both identified a positive correlation between ACC/AHA lesion class and clinical outcomes [4949. Valgimigli M, Serruys PW, Tsuchida K, Vaina S, Morel MA, van den Brand MJ, Colombo A, Morice MC, Dawkins K, de Bruyne B, Kornowski R, de Servi S, Guagliumi G, Jukema JW, Mohr FW, Kappetein AP, Wittebols K, Stoll HP, Boersma E, Parrinello G. 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(8):1072-81. ] [5050. Capodanno D, Di Salvo ME, Cincotta G, Miano M, Tamburino C, Tamburino C. Usefulness of the SYNTAX score for predicting clinical outcome after percutaneous coronary intervention of unprotected left main coronary artery disease. Circ Cardiovasc Interv. 2009;2(4):302-8. ]. Valgmigli et al reported a positive correlation between ACC/AHA lesion score and clinical outcomes amongst 306 patients enrolled in the ARTS-II study who had triple vessel disease and were treated by siriolimus-eluting stents. Similarly, Capodanno et al demonstrated that the ACC/AHA lesion score predicted both cardiac death (p=0.001) and MACCE (p=0.02) at 1-year follow-up amongst 255 patients with LM undergoing PCI with DES [4141. Ranucci M, Castelvecchio S, Conte M, Megliola G, Speziale G, Fiore F, Guarracino F, Scolletta S, Escobar RM, Falco M, Bignami E, Landoni G. The easier, the better: age, creatinine, ejection fraction score for operative mortality risk stratification in a series of 29,659 patients undergoing elective cardiac surgery. J Thorac Cardiovasc Surg. 2011;142(3):581-6. ].

Part of the reason for this discrepancy may relate to the reproducibility of the ACC/AHA lesion score, which has been reported to be good when differentiating Type C from non-Type C lesions (kappa 0.85), however levels of agreement are much poorer when separating Type A, from B and C lesions (kappa 0.48) [4646. Kleiman NS, Rodriguez AR, Raizner AE. Interobserver variability in grading of coronary arterial narrowings using the American College of Cardiology/American Heart Association grading criteria. Am J Cardiol. 1992;69(4):413-5. ].

SYNTAX Score

The SYNTAX score (SXscore) is an angiographic scoring system which allows the complexity of CAD to be quantified [2424. Sianos G, Morel MA, Kappetein AP, Morice MC, Colombo A, Dawkins K, van den Brand M, Van Dyck N, Russell ME, Mohr FW, Serruys PW. The SYNTAX Score: an angiographic tool grading the complexity of coronary artery disease. EuroIntervention. 2005;1(2):219-27. , 2525. Serruys PW, Onuma Y, Garg S, Sarno G, van den Brand M, Kappetein AP, Van Dyck N, Mack M, Holmes D, Feldman T, Morice MC, Colombo A, Bass E, Leadley K, Dawkins KD, van Es GA, Morel MA, Mohr FW. Assessment of the SYNTAX score in the Syntax study. EuroIntervention. 2009;5(1):50-6. ]. Both lesion location ( Figure 1 ) and adverse lesion characteristics ( Figure 1 ) are used to calculate the score, which can be performed using either a downloadable calculator or the SXscore website (https://syntaxscore2020.com/). The score represents an amalgamation of a number of historical anatomical scores such as the AHA grading committee classification of the coronary tree segments (modified for the ARTS study), the modified Leaman score, the ACC/AHA lesion classification, the total occlusion classification, and the Medina bifurcation classification. [5151. Austen WG, Edwards JE, Frye RL, Gensini GG, Gott VL, Griffith LS, McGoon DC, Murphy ML, Roe BB. A reporting system on patients evaluated for coronary artery disease. Report of the Ad Hoc Committee for Grading of Coronary Artery Disease, Council on Cardiovascular Surgery, American Heart Association. Circulation. 1975;51(4 Suppl):5-40. , 5252. Leaman DM, Brower RW, Meester GT, Serruys P, van den Brand M. Coronary artery atherosclerosis: severity of the disease, severity of angina pectoris and compromised left ventricular function. Circulation. 1981;63(2):285-99. , 5353. Hamburger JN, Gijsbers GH, Ozaki Y, Ruygrok PN, de Feyter PJ, Serruys PW. Recanalization of chronic total coronary occlusions using a laser guide wire: a pilot study. J Am Coll Cardiol. 1997;30(3):649-56. , 5454. Smith SC, Jr, Dove JT, Jacobs AK, Kennedy JW, Kereiakes D, Kern MJ, Kuntz RE, Popma JJ, Schaff HV, Williams DO, Gibbons RJ, Alpert JP, Eagle KA, Faxon DP, Fuster V, Gardner TJ, Gregoratos G, Russell RO, Smith SC, Jr. ACC/AHA guidelines for percutaneous coronary intervention (revision of the 1993 PTCA guidelines)-executive summary: a report of the American College of Cardiology/American Heart Association task force on practice guidelines (Committee to revise the 1993 guidelines for percutaneous transluminal coronary angioplasty) endorsed by the Society for Cardiac Angiography and Interventions. Circulation. 2001;103(24):3019-41. , 5555. Medina A, Suárez de Lezo J, Pan M. [A new classification of coronary bifurcation lesions]. Rev Esp Cardiol 2006;59(2):183. ]

The score was devised primarily for use in the SYNTAX trial as a method of bringing together the cardiologist and cardiac surgeon to study in great detail the coronary angiograms of patients enrolled in the trial. On the basis of this assessment those patients considered to have complex disease suitable for treatment with PCI or CABG were entered into the randomised arm of the study, whilst those patients thought to be suitable for only PCI or CABG were entered in the respective PCI and CABG registries. Although not specifically designed to stratify risk, it was hypothesised that the SXscore may correlate with clinical outcomes including procedural risk [2424. Sianos G, Morel MA, Kappetein AP, Morice MC, Colombo A, Dawkins K, van den Brand M, Van Dyck N, Russell ME, Mohr FW, Serruys PW. The SYNTAX Score: an angiographic tool grading the complexity of coronary artery disease. EuroIntervention. 2005;1(2):219-27. ].

The score was first used prospectively in the SYNTAX trial, and has since been tested in a number of different clinical trials both in elective and acute patients, with simple and/or complex disease, followed-up for between 1- and 10-years [1515. Capodanno D, Capranzano P, Di Salvo ME, Caggegi A, Tomasello D, Cincotta G, Miano M, Patané M, Tamburino C, Tolaro S, Patané L, Calafiore AM, Tamburino C. Usefulness of SYNTAX score to select patients with left main coronary artery disease to be treated with coronary artery bypass graft. JACC Cardiovasc Interv. 2009;2(8):731-8. , 1616. Serruys PW, Morice MC, Kappetein AP, Colombo A, Holmes DR, Mack MJ, Ståhle E, Feldman TE, van den Brand M, Bass EJ, Van Dyck N, Leadley K, Dawkins KD, Mohr FW. Percutaneous coronary intervention versus coronary-artery bypass grafting for severe coronary artery disease. N Engl J Med. 2009;360(10):961-72. , 3535. Serruys PW, Onuma Y, Garg S, Vranckx P, De Bruyne B, Morice MC, Colombo A, Macaya C, Richardt G, Fajadet J, Hamm C, Schuijer M, Rademaker T, Wittebols K, Stoll HP. 5-year clinical outcomes of the ARTS II (Arterial Revascularization Therapies Study II) of the sirolimus-eluting stent in the treatment of patients with multivessel de novo coronary artery lesions. J Am Coll Cardiol. 2010;55(11):1093-101. , 3636. Witberg, G., et al., The Prognostic Effects of Coronary Disease Severity and Completeness of Revascularization on Mortality in Patients Undergoing Transcatheter Aortic Valve Replacement. JACC Cardiovasc Interv, 2017. 10(14): p. 1428-1435. , 3737. Onuma Y, Girasis C, Piazza N, Garcia-Garcia HM, Kukreja N, Garg S, Eindhoven J, Cheng JM, Valgimigli M, van Domburg R, Serruys PW. Long-term clinical results following stenting of the left main stem: insights from RESEARCH (Rapamycin-Eluting Stent Evaluated at Rotterdam Cardiology Hospital) and T-SEARCH (Taxus-Stent Evaluated at Rotterdam Cardiology Hospital) Registries. JACC Cardiovasc Interv. 2010;3(6):584-94. , 4949. Valgimigli M, Serruys PW, Tsuchida K, Vaina S, Morel MA, van den Brand MJ, Colombo A, Morice MC, Dawkins K, de Bruyne B, Kornowski R, de Servi S, Guagliumi G, Jukema JW, Mohr FW, Kappetein AP, Wittebols K, Stoll HP, Boersma E, Parrinello G. 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(8):1072-81. , 5050. Capodanno D, Di Salvo ME, Cincotta G, Miano M, Tamburino C, Tamburino C. Usefulness of the SYNTAX score for predicting clinical outcome after percutaneous coronary intervention of unprotected left main coronary artery disease. Circ Cardiovasc Interv. 2009;2(4):302-8. , 5656. Wykrzykowska JJ, Garg S, Girasis C, de Vries T, Morel MA, van Es GA, Buszman P, Linke A, Ischinger T, Klauss V, Corti R, Eberli F, Wijns W, Morice MC, di Mario C, van Geuns RJ, Juni P, Windecker S, Serruys PW. 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(4):272-7. , 5757. Girasis C, Garg S, Räber L, Sarno G, Morel MA, Garcia-Garcia HM, Lüscher TF, Serruys PW, Windecker S. SYNTAX score and Clinical SYNTAX score as predictors of very long-term clinical outcomes in patients undergoing percutaneous coronary interventions: a substudy of SIRolimus-eluting stent compared with pacliTAXel-eluting stent for coronary revascularization (SIRTAX) trial. Eur Heart J. 2011;32(24):3115-27. , 5858. Garg S, Sarno G, Serruys PW, Rodriguez AE, Bolognese L, Anselmi M, De Cesare N, Colangelo S, Moreno R, Gambetti S, Monti M, Bristot L, Bressers M, Garcia-Garcia HM, Parrinello G, Campo G, Valgimigli M. Prediction of 1-year clinical outcomes using the SYNTAX score in patients with acute ST-segment elevation myocardial infarction undergoing primary percutaneous coronary intervention: a substudy of the STRATEGY (Single High-Dose Bolus Tirofiban and Sirolimus-Eluting Stent Versus Abciximab and Bare-Metal Stent in Acute Myocardial Infarction) and MULTISTRATEGY (Multicenter Evaluation of Single High-Dose Bolus Tirofiban Versus Abciximab With Sirolimus-Eluting Stent or Bare-Metal Stent in Acute Myocardial Infarction Study) trials. JACC Cardiovasc Interv. 2011;4(1):66-75. , 5959. Garg S, Serruys PW, Silber S, Wykrzykowska J, van Geuns RJ, Richardt G, Buszman PE, Kelbæk H, van Boven AJ, Hofma SH, Linke A, Klauss V, Wijns W, Macaya C, Garot P, DiMario C, Manoharan G, Kornowski R, Ischinger T, Bartorelli A, Van Remortel E, Ronden J, Windecker S. The prognostic utility of the SYNTAX score on 1-year outcomes after revascularization with zotarolimus- and everolimus-eluting stents: a substudy of the RESOLUTE All Comers Trial. JACC Cardiovasc Interv. 2011;4(4):432-41. , 6060. Thuijs D, Kappetein AP, Serruys PW, Mohr FW, Morice MC, Mack MJ, Holmes DR, Jr, Curzen N, Davierwala P, Noack T, Milojevic M, Dawkins KD, da Costa BR, Juni P, Head SJ. Percutaneous coronary intervention versus coronary artery bypass grafting in patients with three-vessel or left main coronary artery disease: 10-year follow-up of the multicentre randomised controlled SYNTAX trial. Lancet. 2019;394(10206):1325-1334. ]. In all studies irrespective of follow-up duration, or clinical presentation, a higher SXscore tercile has consistently been associated with the poorest outcomes [1515. Capodanno D, Capranzano P, Di Salvo ME, Caggegi A, Tomasello D, Cincotta G, Miano M, Patané M, Tamburino C, Tolaro S, Patané L, Calafiore AM, Tamburino C. Usefulness of SYNTAX score to select patients with left main coronary artery disease to be treated with coronary artery bypass graft. JACC Cardiovasc Interv. 2009;2(8):731-8. , 1616. Serruys PW, Morice MC, Kappetein AP, Colombo A, Holmes DR, Mack MJ, Ståhle E, Feldman TE, van den Brand M, Bass EJ, Van Dyck N, Leadley K, Dawkins KD, Mohr FW. Percutaneous coronary intervention versus coronary-artery bypass grafting for severe coronary artery disease. N Engl J Med. 2009;360(10):961-72. , 3535. Serruys PW, Onuma Y, Garg S, Vranckx P, De Bruyne B, Morice MC, Colombo A, Macaya C, Richardt G, Fajadet J, Hamm C, Schuijer M, Rademaker T, Wittebols K, Stoll HP. 5-year clinical outcomes of the ARTS II (Arterial Revascularization Therapies Study II) of the sirolimus-eluting stent in the treatment of patients with multivessel de novo coronary artery lesions. J Am Coll Cardiol. 2010;55(11):1093-101. , 3636. Witberg, G., et al., The Prognostic Effects of Coronary Disease Severity and Completeness of Revascularization on Mortality in Patients Undergoing Transcatheter Aortic Valve Replacement. JACC Cardiovasc Interv, 2017. 10(14): p. 1428-1435. , 3737. Onuma Y, Girasis C, Piazza N, Garcia-Garcia HM, Kukreja N, Garg S, Eindhoven J, Cheng JM, Valgimigli M, van Domburg R, Serruys PW. Long-term clinical results following stenting of the left main stem: insights from RESEARCH (Rapamycin-Eluting Stent Evaluated at Rotterdam Cardiology Hospital) and T-SEARCH (Taxus-Stent Evaluated at Rotterdam Cardiology Hospital) Registries. JACC Cardiovasc Interv. 2010;3(6):584-94. , 4949. Valgimigli M, Serruys PW, Tsuchida K, Vaina S, Morel MA, van den Brand MJ, Colombo A, Morice MC, Dawkins K, de Bruyne B, Kornowski R, de Servi S, Guagliumi G, Jukema JW, Mohr FW, Kappetein AP, Wittebols K, Stoll HP, Boersma E, Parrinello G. 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(8):1072-81. , 5050. Capodanno D, Di Salvo ME, Cincotta G, Miano M, Tamburino C, Tamburino C. Usefulness of the SYNTAX score for predicting clinical outcome after percutaneous coronary intervention of unprotected left main coronary artery disease. Circ Cardiovasc Interv. 2009;2(4):302-8. , 5656. Wykrzykowska JJ, Garg S, Girasis C, de Vries T, Morel MA, van Es GA, Buszman P, Linke A, Ischinger T, Klauss V, Corti R, Eberli F, Wijns W, Morice MC, di Mario C, van Geuns RJ, Juni P, Windecker S, Serruys PW. 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(4):272-7. , 5757. Girasis C, Garg S, Räber L, Sarno G, Morel MA, Garcia-Garcia HM, Lüscher TF, Serruys PW, Windecker S. SYNTAX score and Clinical SYNTAX score as predictors of very long-term clinical outcomes in patients undergoing percutaneous coronary interventions: a substudy of SIRolimus-eluting stent compared with pacliTAXel-eluting stent for coronary revascularization (SIRTAX) trial. Eur Heart J. 2011;32(24):3115-27. , 5959. Garg S, Serruys PW, Silber S, Wykrzykowska J, van Geuns RJ, Richardt G, Buszman PE, Kelbæk H, van Boven AJ, Hofma SH, Linke A, Klauss V, Wijns W, Macaya C, Garot P, DiMario C, Manoharan G, Kornowski R, Ischinger T, Bartorelli A, Van Remortel E, Ronden J, Windecker S. The prognostic utility of the SYNTAX score on 1-year outcomes after revascularization with zotarolimus- and everolimus-eluting stents: a substudy of the RESOLUTE All Comers Trial. JACC Cardiovasc Interv. 2011;4(4):432-41. ] , with several studies also identifying the SXscore to be an independent predictor of MACCE [1515. Capodanno D, Capranzano P, Di Salvo ME, Caggegi A, Tomasello D, Cincotta G, Miano M, Patané M, Tamburino C, Tolaro S, Patané L, Calafiore AM, Tamburino C. Usefulness of SYNTAX score to select patients with left main coronary artery disease to be treated with coronary artery bypass graft. JACC Cardiovasc Interv. 2009;2(8):731-8. , 3535. Serruys PW, Onuma Y, Garg S, Vranckx P, De Bruyne B, Morice MC, Colombo A, Macaya C, Richardt G, Fajadet J, Hamm C, Schuijer M, Rademaker T, Wittebols K, Stoll HP. 5-year clinical outcomes of the ARTS II (Arterial Revascularization Therapies Study II) of the sirolimus-eluting stent in the treatment of patients with multivessel de novo coronary artery lesions. J Am Coll Cardiol. 2010;55(11):1093-101. , 3636. Witberg, G., et al., The Prognostic Effects of Coronary Disease Severity and Completeness of Revascularization on Mortality in Patients Undergoing Transcatheter Aortic Valve Replacement. JACC Cardiovasc Interv, 2017. 10(14): p. 1428-1435. , 4949. Valgimigli M, Serruys PW, Tsuchida K, Vaina S, Morel MA, van den Brand MJ, Colombo A, Morice MC, Dawkins K, de Bruyne B, Kornowski R, de Servi S, Guagliumi G, Jukema JW, Mohr FW, Kappetein AP, Wittebols K, Stoll HP, Boersma E, Parrinello G. 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(8):1072-81. , 5050. Capodanno D, Di Salvo ME, Cincotta G, Miano M, Tamburino C, Tamburino C. Usefulness of the SYNTAX score for predicting clinical outcome after percutaneous coronary intervention of unprotected left main coronary artery disease. Circ Cardiovasc Interv. 2009;2(4):302-8. , 5656. Wykrzykowska JJ, Garg S, Girasis C, de Vries T, Morel MA, van Es GA, Buszman P, Linke A, Ischinger T, Klauss V, Corti R, Eberli F, Wijns W, Morice MC, di Mario C, van Geuns RJ, Juni P, Windecker S, Serruys PW. 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(4):272-7. , 5858. Garg S, Sarno G, Serruys PW, Rodriguez AE, Bolognese L, Anselmi M, De Cesare N, Colangelo S, Moreno R, Gambetti S, Monti M, Bristot L, Bressers M, Garcia-Garcia HM, Parrinello G, Campo G, Valgimigli M. Prediction of 1-year clinical outcomes using the SYNTAX score in patients with acute ST-segment elevation myocardial infarction undergoing primary percutaneous coronary intervention: a substudy of the STRATEGY (Single High-Dose Bolus Tirofiban and Sirolimus-Eluting Stent Versus Abciximab and Bare-Metal Stent in Acute Myocardial Infarction) and MULTISTRATEGY (Multicenter Evaluation of Single High-Dose Bolus Tirofiban Versus Abciximab With Sirolimus-Eluting Stent or Bare-Metal Stent in Acute Myocardial Infarction Study) trials. JACC Cardiovasc Interv. 2011;4(1):66-75. , 5959. Garg S, Serruys PW, Silber S, Wykrzykowska J, van Geuns RJ, Richardt G, Buszman PE, Kelbæk H, van Boven AJ, Hofma SH, Linke A, Klauss V, Wijns W, Macaya C, Garot P, DiMario C, Manoharan G, Kornowski R, Ischinger T, Bartorelli A, Van Remortel E, Ronden J, Windecker S. The prognostic utility of the SYNTAX score on 1-year outcomes after revascularization with zotarolimus- and everolimus-eluting stents: a substudy of the RESOLUTE All Comers Trial. JACC Cardiovasc Interv. 2011;4(4):432-41. ] and/or mortality [3737. Onuma Y, Girasis C, Piazza N, Garcia-Garcia HM, Kukreja N, Garg S, Eindhoven J, Cheng JM, Valgimigli M, van Domburg R, Serruys PW. Long-term clinical results following stenting of the left main stem: insights from RESEARCH (Rapamycin-Eluting Stent Evaluated at Rotterdam Cardiology Hospital) and T-SEARCH (Taxus-Stent Evaluated at Rotterdam Cardiology Hospital) Registries. JACC Cardiovasc Interv. 2010;3(6):584-94. , 5050. Capodanno D, Di Salvo ME, Cincotta G, Miano M, Tamburino C, Tamburino C. Usefulness of the SYNTAX score for predicting clinical outcome after percutaneous coronary intervention of unprotected left main coronary artery disease. Circ Cardiovasc Interv. 2009;2(4):302-8. , 5656. Wykrzykowska JJ, Garg S, Girasis C, de Vries T, Morel MA, van Es GA, Buszman P, Linke A, Ischinger T, Klauss V, Corti R, Eberli F, Wijns W, Morice MC, di Mario C, van Geuns RJ, Juni P, Windecker S, Serruys PW. 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(4):272-7. , 5858. Garg S, Sarno G, Serruys PW, Rodriguez AE, Bolognese L, Anselmi M, De Cesare N, Colangelo S, Moreno R, Gambetti S, Monti M, Bristot L, Bressers M, Garcia-Garcia HM, Parrinello G, Campo G, Valgimigli M. Prediction of 1-year clinical outcomes using the SYNTAX score in patients with acute ST-segment elevation myocardial infarction undergoing primary percutaneous coronary intervention: a substudy of the STRATEGY (Single High-Dose Bolus Tirofiban and Sirolimus-Eluting Stent Versus Abciximab and Bare-Metal Stent in Acute Myocardial Infarction) and MULTISTRATEGY (Multicenter Evaluation of Single High-Dose Bolus Tirofiban Versus Abciximab With Sirolimus-Eluting Stent or Bare-Metal Stent in Acute Myocardial Infarction Study) trials. JACC Cardiovasc Interv. 2011;4(1):66-75. ] in patients undergoing PCI. These findings have been confirmed in a patient level pooled analysis of seven contemporary drug-eluting stent trials (SYNTAX, RESOLUTE, LEADERS, SIRTAX, STRATEGY, MULTI-STRATEGY, and ARTS-II) which stratified 1-year outcomes from over 6,500 patients according to quartiles of the SXscore [6161. Garg S, Sarno G, Girasis C, Vranckx P, de Vries T, Swart M, Bressers M, Garcia-Garcia HM, van Es GA, Räber L, Campo G, Valgimigli M, Dawkins KD, Windecker S, Serruys PW. A patient-level pooled analysis assessing the impact of the SYNTAX (synergy between percutaneous coronary intervention with taxus and cardiac surgery) score on 1-year clinical outcomes in 6,508 patients enrolled in contemporary coronary stent trials. JACC Cardiovasc Interv. 2011;4(6):645-53. ] . Overall these results support a role for the SXscore in risk stratifying patients following diagnostic coronary angiography, which has been formally recognised in the latest guidelines on myocardial revascularization on both sides of the Atlantic [55. Patel MR, Calhoon JH, Dehmer GJ, Grantham JA, Maddox TM, Maron DJ, Smith PK. ACC/AATS/AHA/ASE/ASNC/SCAI/SCCT/STS 2016 Appropriate Use Criteria for Coronary Revascularization in Patients With Acute Coronary Syndromes: A Report of the American College of Cardiology Appropriate Use Criteria Task Force, American Association for Thoracic Surgery, American Heart Association, American Society of Echocardiography, American Society of Nuclear Cardiology, Society for Cardiovascular Angiography and Interventions, Society of Cardiovascular Computed Tomography, and the Society of Thoracic Surgeons. J Am Coll Cardiol. 2017;69(5):570-591. , 1414. Neumann FJ, Sousa-Uva M, Ahlsson A, Alfonso F, Banning AP, Benedetto U, Byrne RA, Collet JP, Falk V, Head SJ, Juni P, Kastrati A, Koller A, Kristensen SD, Niebauer J, Richter DJ, Seferovic PM, Sibbing D, Stefanini GG, Windecker S, Yadav R, Zembala MO, Group ESCSD. 2018 ESC/EACTS Guidelines on myocardial revascularization. Eur Heart J. 2019;40(2):87-165. ].

In the SYNTAX study, in those patients treated with CABG, MACCE occurred in 26.9% compared with 37.3% in those treated with PCI (p<0.0001) at 5 years of follow-up. While all cause death and stroke were not significantly different between CABG and PCI, rates of MI and repeat revascularization were significantly less with CABG. The tertile of SYNTAX Score at baseline had a significant association with MACCE ( Figure 2). As the angiographic extent and complexity of disease increased, the difference in MACCE rates between patients randomized to CABG and those to PCI also increased [6262. Mohr FW, Morice MC, Kappetein AP, Feldman TE, Stahle E, Colombo A, Mack MJ, Holmes DR, Jr, Morel MA, Van Dyck N, Houle VM, Dawkins KD, Serruys PW. Coronary artery bypass graft surgery versus percutaneous coronary intervention in patients with three-vessel disease and left main coronary disease: 5-year follow-up of the randomised, clinical SYNTAX trial. Lancet. 2013;381(9867):629-38. ] . In the SYNTAXES study, 10-year follow-up was performed. Overall, 10-year mortality was 28% in the PCI arm and 24% in the CABG arm (p=0.066) ( Figure 3) [6060. Thuijs D, Kappetein AP, Serruys PW, Mohr FW, Morice MC, Mack MJ, Holmes DR, Jr, Curzen N, Davierwala P, Noack T, Milojevic M, Dawkins KD, da Costa BR, Juni P, Head SJ. Percutaneous coronary intervention versus coronary artery bypass grafting in patients with three-vessel or left main coronary artery disease: 10-year follow-up of the multicentre randomised controlled SYNTAX trial. Lancet. 2019;394(10206):1325-1334. ]. Although the mortality rates of patients with SYNTAX Score ≤ 32 were not significantly different in PCI and CABG arms, patients with SYNTAX Score ≥ 33 received a benefit from CABG (HR: 1.47; 95% CI [1.10-1.96]).

These data suggest that the SXscore has two main roles in clinical practice. Firstly, the SXscore can assist in risk stratifying patients undergoing percutaneous revascularization. In addition to this, outcomes from studies which have included a surgical treatment arm indicate the SXscore also has a utility in assisting clinicians with important revascularization decisions.

The SYNTAX score has also been used to assess the importance of implementing a specific revascularization strategy. As previously mentioned, a residual SYNTAX score has been developed and studied in the PCI cohort of the SYNTAX study. Using this score, following PCI, patients could be categorized as having been completely revascularized, or having a variable degree of incomplete revascularization. In this study, complete revascularization was obtained in 42.7% of patients. In the remainder, revascularization was incomplete. Based on the tertile of residual score (>0-4, >4-8, and >8), during 5 years of follow-up there was a graded increase in subsequent events. Patients with a residual score >8 had a significantly higher risk of 10-year all-cause death as compared with those undergoing PCI with complete revascularization (50.1% vs. 22.2%; adjusted HR:3.40; 95% CI: 2.13-5.43) [6363. Takahashi K, Serruys PW, Gao C, Ono M, Wang R, Thuijs D, Mack MJ, Curzen N, Mohr FW, Davierwala P, Milojevic M, Wykrzykowska JJ, de Winter RJ, Sharif F, Onuma Y, Head SJ, Kappetein AP, Morice MC, Holmes DR, Jr, Investigators SESS. Ten-Year All-Cause Death According to Completeness of Revascularization in Patients With Three-Vessel Disease or Left Main Coronary Artery Disease: Insights From the SYNTAX Extended Survival Study. Circulation. 2021;144(2):96-109. ] . This has important implications for clinical practice. In patients with MVD in whom only incomplete revascularization is expected to be achieved with PCI, surgery should be more strongly considered if possible as a more optimal revascularization strategy.

More recently the residual SYNTAX score has been assessed for its prognostic impact in patients with cardiogenic shock. In the The CULPRIT-SHOCK trial, residual SYNTAX score was independently associated with 30-day mortality (adjusted odds ratio per 10 units: 1.49; 95% CI: 1.11 to 2.01) and 1-year mortality (adjusted odds ratio per 10 units: 1.52; 95% CI: 1.11 to 2.07) [6464. Barthélémy O, Rouanet S, Brugier D, Vignolles N, Bertin B, Zeitouni M, Guedeney P, Hauguel-Moreau M, Hage G, Overtchouk P, Akin I, Desch S, Vicaut E, Zeymer U, Thiele H, Montalescot G. Predictive Value of the Residual SYNTAX Score in Patients With Cardiogenic Shock. J Am Coll Cardiol. 2021;77(2):144-155. ].

COMBINED RISK SCORES

Current data indicates that the predictive ability of clinical- and angiographic-based risk models is partly dependent on the endpoint being assessed. For example, angiographic-based models appear to have an inferior predictive ability when compared to clinical-based models for the assessment of “hard clinical” endpoints such as mortality, which is a consequence of mortality being heavily influenced by pre-morbid comorbidities. By contrast angiographic-based models appear to be superior in the prediction of “softer” endpoints such as repeat revascularization. Taking this into consideration has prompted the development of newer risk models which seek to combine clinical and angiographic variables in a single model with the aim of ultimately developing the most effective overall method of risk stratification. These newer models are in their infancy, however initial results are encouraging.

EuroSCORE-SYNTAX

The EuroSCORE and the SXscore are presently the most extensively studied risk models in patients undergoing revascularization. Combining both models potentially offers the prospect of harnessing the positive aspects of each, namely the ability of the EuroSCORE to identify patients at high-risk of adverse events irrespective of treatment modality, and the ability of the SXscore to stratify risk in those undergoing PCI, and assist in the selection of optimal revascularization strategy.

Despite this potential, there have been some practical difficulties in establishing the optimal method of combining the two models. A simple method tried in the SYNTAX study involved sub-dividing patients in the SXscore tertiles by a EuroSCORE above or below the median. Although in principal this was appealing, it failed to translate into a consistent and understandable relationship, which may partly have been the consequence of a small numbers of patients in each sub-group.

The most promising method of combining both models to-date has been described by Capodanno et al, who developed the Global Risk Classification (GRC) [1212. Capodanno D, Miano M, Cincotta G, Caggegi A, Ruperto C, Bucalo R, Sanfilippo A, Capranzano P, Tamburino C. EuroSCORE refines the predictive ability of SYNTAX score in patients undergoing left main percutaneous coronary intervention. Am Heart J. 2010;159(1):103-9. ] . This model categorises patients into low, medium and high-risk using a matrix which incorporates a patient’s EuroSCORE, which is sub-divided into the historically defined groups of low (0-2), intermediate (3-5) and high-risk (≥6), and their SXscore, which is divided into low, intermediate and high terciles ( Box 1 ). In the initial study, the GRC was assessed in 255 patients from the LM registry in whom SXscores were calculated retrospectively. At 2-years follow-up rates of cardiac death were 1.6%, 16.0% and 31.4% in the respective low, intermediate and high GRC groups. Most importantly, the GRC was shown to have a superior discriminatory ability when compared with other risk scores, including the EuroSCORE and the SXscore, for the prediction of in-hospital and 2-year mortality. In the SYNTAX trial, the GRC substantially enhanced the identification of low-risk patients who could safely and efficaciously be treated with CABG or PCI, compared to the SXscore [6565. Serruys PW, Farooq V, Vranckx P, Girasis C, Brugaletta S, Garcia-Garcia HM, Holmes DR, Jr, Kappetein AP, Mack MJ, Feldman T, Morice MC, Stahle E, James S, Colombo A, Pereda P, Huang J, Morel MA, Van Es GA, Dawkins KD, Mohr FW, Steyerberg EW. A global risk approach to identify patients with left main or 3-vessel disease who could safely and efficaciously be treated with percutaneous coronary intervention: the SYNTAX Trial at 3 years. JACC Cardiovasc Interv. 2012;5(6):606-17. ].

FOCUS BOX 1

The Global Risk Classification matrix [1212. Capodanno D, Miano M, Cincotta G, Caggegi A, Ruperto C, Bucalo R, Sanfilippo A, Capranzano P, Tamburino C. EuroSCORE refines the predictive ability of SYNTAX score in patients undergoing left main percutaneous coronary intervention. Am Heart J. 2010;159(1):103-9. ]

• Global Risk Classification (GRC) is derived using the above matrix
• The GRC divides patients into Low, Intermediate and High-risk groups as shown

An additional score from the SYNTAX trial included the addition of age, creatinine clearance and ejection fraction as a variable. To this were added peripheral vascular disease, chronic obstructive pulmonary disease, presence of unprotected left main coronary artery stenosis and female sex.

The model (SYNTAX score II) was developed to predict 4-year all-cause death in all 1800 patients randomized in the initial SYNTAX trial. Baseline age, creatinine clearance, ejection fraction, peripheral vascular disease, chronic obstructive pulmonary disease, presence of unprotected left main coronary artery stenosis and female sex are applied in addition to the SYNTAX score. The model was then externally validated using data from the multinational non-randomized all-comers DELTA registry of 2,891 patients with unprotected left main stenosis either alone or in combination with single or multivessel coronary disease [3434. Roques F, Michel P, Goldstone AR, Nashef SA. The logistic EuroSCORE. Eur Heart J. 2003;24(9):881-2. ]. Farooq et al. found that the SYNTAX score II discriminated well in both CABG and PCI patients. The concordance index for internal validation was 0.725 while for external validation was 0.716, which was substantially higher than the anatomical SXscore along (C-indices= 0.567 and 0.612, respectively). Accordingly, the authors concluded that, with the score based on both angiographic and selected clinical variables, can provide valuable information for decision making and recommendations of optimal strategies between CABG and PCI in patients with more complex and extensive coronary artery disease.

After that, based on the longer follow-up in the SYNTAX(ES) trial, the SYNTAX score II 2020 to predict 5-year risk of MACCE and 10-year risk of all-cause mortality depending on whether the patient had undergone PCI or CABG was (re)developed. For MACCE, a composite of all cause death, non-fatal stroke or non-fatal myocardial infarction was used ( Figure 4). This score was then validated using patient-level data from the FREEDOM, BEST, PRECOMBAT and EXCEL trials all of which evaluated patients with multivessel or left main coronary artery disease undergoing either PCI or CABG.

10-year all-cause mortality:

At 10 years, a total of 460 deaths occurred among all 1,800 patients. In the entire group, there was no significant difference in mortality at 10 years between PCI (28%) and CABG (24%) and (HR: 1.19, 95% CI: 0.99-1.43, log-rank p= 0.066). The risk predictive model included 8 clinical factors. When the initial revascularization strategy was added in, the effect of initial procedure was an important factor. There was a beneficial effect of CABG in patients treated for three-vessel disease (HR 0.67, 95%CI 0,53-0.86). This benefit was not seen in patients with LMCA disease (HR 0.92, 95% CI 0.72-1.19).

The SYNTAX Score II 2020 had good discrimination ability for both PCI and CABG (C-index 0.73 for both groups. The estimated treatment benefit of CABG over PCI varied substantially among patients in the SYNTAXES trial, and the benefit predictions were well calibrated ( Figure 5). [6666. Takahashi K, Serruys PW, Fuster V, Farkouh ME, Spertus JA, Cohen DJ, Park SJ, Park DW, Ahn JM, Kappetein AP, Head SJ, Thuijs DJFM, Onuma Y, Kent DM, Steyerberg EW, van Klaveren D, Investigator ST, Investigator FT, Investigator BT, Investigator PT. Redevelopment and validation of the SYNTAX score II to individualise decision making between percutaneous and surgical revascularisation in patients with complex coronary artery disease: secondary analysis of the multicentre randomised controlled SYNTAXES trial with external cohort validation. Lancet. 2020;396(10260):1399-1412. ] In the CREDO-Kyoto registry, the SYNTAX Score II 2020 for 5-year mortality well predicted the prognosis after PCI and CABG. Predicted absolute risk difference (predicted mortality rate after PCI - predicted mortality rate after CABG) of <4.5% and ≥4.5% can offer sensible treatment recommendations of either “equipoise of PCI and CABG” or “CABG better”, respectively. [6767. Hara H, Shiomi H, van Klaveren D, Kent DM, Steyerberg EW, Garg S, Onuma Y, Kimura T, Serruys PW. External validation of the SYNTAX Score II 2020. J Am Coll Cardiol. 2021. ]

5-year MACCE:

At 5 years, 21% of the PCI group had a MACCE endpoint versus 17% of CABG patients (HR 1.27, 95% CI 1.03-1.59, p = 0.03). The model for MACCE uses the same 8 clinical factors as the model for death, and demonstrated helpful discrimination both in PCI and CABG patients (C-index 0.65 and 0.71, respectively). The SYNTAX score II 2020 for 5-year MACCE was externally validated using the FREEDOM, BEST, PRECOMBAT trials, and showed helpful discrimination for PCI (C-index=0.67) and CABG (C-index=0.62). Subsequently, the EXCEL trial was included in the external validation (the FREEDOM, BEST, PRECOMBAT, and EXCEL trials). Its discrimination was also helpful (PCI: C-index=0.65; CABG: C-index =0.61), and the benefit predictions were well calibrated ( Figure 6).[6868. Takahashi K, van Klaveren D, Steyerberg EW, Onuma Y, Serruys PW. Concerns with the new SYNTAX score - Authors’ reply. Lancet. 2021;397(10276):795-796. ]

Overall, these studies reiterate the importance of considering both clinical and angiographic variables in the assessment of overall risk, and provided a combined scoring system which on initial assessment appears to hold promise for the future.

MODELS FOR BOTH BLEEDING AND ISCHEMIC RISK

PARIS score

The PARIS score was developed from patients treated with percutaneous coronary intervention (PCI) regardless of their clinical presentations, and predict out-of-hospital BARC 3 or 5 bleeding and thrombotic risk (MI and/or definite or probable stent thrombosis) ( Figure 7). [6969. Baber U, Mehran R, Giustino G, Cohen DJ, Henry TD, Sartori S, Ariti C, Litherland C, Dangas G, Gibson CM, Krucoff MW, Moliterno DJ, Kirtane AJ, Stone GW, Colombo A, Chieffo A, Kini AS, Witzenbichler B, Weisz G, Steg PG, Pocock S. Coronary Thrombosis and Major Bleeding After PCI With Drug-Eluting Stents: Risk Scores From PARIS. J Am Coll Cardiol. 2016;67(19):2224-2234. ] A total of 10 variables (9 baseline characteristics and 1 medication status at discharge) is applied to risk prediction. In the development and validation cohorts, C-statistics were 0.72 and 0.64 for major bleeding, and 0.70 and 0.65 for thrombotic events, respectively.

ARC-high bleeding risk (HBR) trade-off model

This score provides the predicted 1-year risk of non-periprocedural major bleeding (BARC 3-5 bleeding) and thrombotic events (MI and/or definite or probable stent thrombosis) after coronary stenting in HBR patients ( Figure 7). Nine baseline characteristics, 2 procedural variables, and 1 medication status at discharge are used for score calculation. In the development and validation cohorts, C-statistics were 0.68 and 0.74 for major bleeding, and 0.68 and 0.74 for thrombotic events, respectively [7070. Urban P, Gregson J, Owen R, Mehran R, Windecker S, Valgimigli M, Varenne O, Krucoff M, Saito S, Baber U, Chevalier B, Capodanno D, Morice MC, Pocock S. Assessing the Risks of Bleeding vs Thrombotic Events in Patients at High Bleeding Risk After Coronary Stent Implantation: The ARC-High Bleeding Risk Trade-off Model. Jama. Cardiol 2021;6(4):410-419. ]. Even though helpful discrimination was demonstrated in the original publication, C-statistics were 0.56 for major bleeding and 0.67 for thrombotic events in the external validation using the database of the GLOBAL LEDERS trial [7171. Hara H, Ono M, Kawashima H, Onuma Y, Serruys PW. Trade-Off Between Bleeding and Thrombotic Risk in Patients With Academic Research Consortium for High Bleeding Risk. Jama Cardiol. 2021. ].

PRAISE score

Machine-learning algorithms have been introduced into the prediction models. The PRAISE score was developed for the prediction of all-cause death, MI, and BARC type 3 or 5 major bleeding 1-year after discharge in ACS patients using a machine-learning algorithm ( Figure 7). In the training, internal validation and external validation, C-indexes were 0.91, 0.82 and 0.92 for death, 0.88, 0.74 and 0.81 for MI, and 0.87, 0.70 and 0.86 for major bleeding, respectively [7272. D’Ascenzo F, De Filippo O, Gallone G, Mittone G, Deriu MA, Iannaccone M, Ariza-Solé A, Liebetrau C, Manzano-Fernández S, Quadri G, Kinnaird T, Campo G, Simao Henriques JP, Hughes JM, Dominguez-Rodriguez A, Aldinucci M, Morbiducci U, Patti G, Raposeiras-Roubin S, Abu-Assi E, De Ferrari GM. Machine learning-based prediction of adverse events following an acute coronary syndrome (PRAISE): a modelling study of pooled datasets. Lancet. 2021;397(10270):199-207. ]. This model needs additional external validation to confirm its utility. Although patients are limited to ACS population, the good predictivities for both bleeding and ischemic events may support the individual optimal antiplatelet and anticoagulant therapy.

These risk models provide both bleeding risk and ischemic risk after PCI. The trade-off between bleeding and ischemic risk can support the optimal antiplatelet and antithrombotic therapy. Therefore, these models should be assessed in daily practice, although the evidences from prospective evaluation of these scores is still warranted.

Conclusions

Risk stratification remains a cornerstone for procedural selection, patient and family counselling and education, and then procedural performance. The field has become more complex because of the number of variables that have been identified as risk factors, which is the consequence of the increasing complexity of patients and their anatomy. The development of the concept of a “Heart Team” comprising of cardiovascular surgeons, and interventional cardiologists and clinical cardiologists working together to select and recommend optimal treatment strategies for a specific patient has become the guideline recommended approach to care. In order to accomplish this, models of risk stratification are essential. A large number of classification schemes have been developed and an increasing amount of information is available on both short-term and now longer-term outcomes in terms of the risk of death, myocardial infarction, stroke and renal damage among others related to specific types of revascularization approaches, surgical versus interventional. These models allow the physicians to personalize care for the specific patient more closely. Selection of the specific model has become more complex because of the number of models available. Cross comparison of the performance of models against each other remains somewhat limited. Some models without angiographic data but just based on clinical parameters have the advantage of being able to be used prior to angiography. More robust models include both angiographic as well as clinical variables. The integration of models in clinical practice with hand held calculators will become increasingly important driven by the mandate to optimize outcome. Inclusion of these models will be an essential part of comparative effectiveness research. A fundamental tenet in this field relates to the fact that the risk stratification model must be parsimonious. A model that includes for example 100 or 150 variables, while it might have some increase in specificity and specificity and accuracy will be of only very limited if any practical clinical utility. The exact number of variable in any predictive model must be as short as able to have clinical value.

Personal perspective - David R. Holmes

Risk stratification schemes and interventional cardiology continue to evolve from qualitative to quantitative approaches. The subject of the specific risk stratification scheme ranges widely from predicting risk of renal failure to myocardial infarction to restenosis (both angiographic and clinical) and mortality, both short and long-term. The specific variables evaluated vary and may include angiographic findings, patient demographics, and clinical presentation and combinations among these. Other important variables include more sophisticated angiographic findings such as FFR, IVUS, OCT. Yet another group of variables not considered have been genetic. For example the issue of differences in response to clopidogrel has been repeatedly emphasized. How these issues may add to longer term risk prediction models of restenosis or stent thrombosis is still the focus of great interest. In addition, it is conceivable because there are different genetic differences in different racial groups, that some risk prediction models may be more robust in certain groups that in others. The number of algorithms continues to proliferate wildly and have been found in validated series to be robust.

Risk scores can be invaluable in determining the risk/benefit of specific therapeutic strategy recommendations. They should be incorporated in the deliberations of the Heart Team and be used in implementing patient centric care decision making. In deciding when and how to incorporate any specific risk score into practice, the Heart Team must be cognizant of multiple issues and challenges. To re-emphasize, to be clinically relevant in actual practice the specific model must be parsimonious and include just those most important variables which are found to make the largest difference in separating our a more desirable/optimal outcome from a much less optimal outcome.

1. The specific risk score must be relevant to the patient at hand.
2. The risk score needs to have been validated in external data sets and be found to be robust.
3. The risk score must include the data which either is available or can be obtained in the specific patient.
4. The risk score should be able to be used at the point of care so that the physician, health care team, and patient receive full benefit.
5. The risk score can only be used as a guideline, because each patient is unique and offer unique challenges as well as unintended consequences.
6. The ideal risk score will never be available, as the medical information used in decision making and the opportunities available continue to evolve. Accordingly risk scores need to continually evolve
7. Finally, the specific score used should be as closely as possible matched to the specific patient being evaluated.