After analyzing the work that we will evaluate in detail below, the conclusions are clear: while prospects look promising, caution is essential. This document, resulting from over a year’s work since the 2021 proposal to revise the 2018 revascularization guidelines. Culminated in the 2022 version after a meeting between societies, published in 2023 due to the extensive drafting process.

Understanding these types of documents requires analyzing the behind-the-scenes work that, in this case, is published with unprecedented transparency. The work stems from a meeting of six members from each cardiological and surgical society, forming a panel of twelve who signed the document. Among cardiologists, both clinicians and interventionists were represented. All panelists’ conflict-of-interest statements were thoroughly reviewed to produce an impartial document, aided by expert statisticians. Statements in the document, presented as a mini-clinical guide, were obtained by a consensus exceeding 75% agreement. Unlike previous documents, such as the 2021 American guidelines, both European societies endorsed the document.

The objective was to establish the optimal treatment for the most debated field in myocardial revascularization. Stable or stabilized left main coronary artery disease (LMCA) after an acute event, with suitable complexity for percutaneous coronary intervention (PCI) or coronary artery bypass grafting (CABG), in relation to a low (0-22 points) or intermediate (23-32 points) SYNTAX score and low surgical risk.

The authors offer a retrospective of the statements made in the 2018 guideline and the changes that motivated the 2022 consensus assertions. The retrospective starts with the debate generated by the 3-year results of the EXCEL study and the well-known controversy over the definition of periprocedural acute myocardial infarction (AMI). This study influenced the 2018 guideline, granting a Class I indication to surgery, while PCI maintained a Class I indication for SYNTAX <22 points, IIa for SYNTAX 23-32, and III for SYNTAX >33. The 2018 guideline also referred to peri-procedural risk assessment, crucial in choosing between therapeutic options, acknowledging the STS score over EuroSCORE II for predicting 30-day surgical mortality, but not for PCI. Conversely, the SYNTAX score correlated well with early mortality post-PCI but did not adequately assess surgical risk.

Since 2018, substantial literature justifying this review has emerged, previously highlighted in earlier blog entries. This includes the updated 5-year results of the EXCEL and NOBLE studies, and the 10-year outcomes of SYNTAX and PRECOMBAT. These are the four comparative studies of PCI vs. CABG in LMCA disease, where drug-eluting stents were used in the percutaneous alternative. The authors also reference a meta-analysis by Sabatine et al., combining data from these studies.

Regarding risk assessment, the authors mention the new SYNTAX II system, which emerged after 2018 due to the poor adjustment of the previous system for surgical risk, incorporating clinical variables. However, it still presented notable interobserver variability in assessing coronary anatomy and overestimated risk in the 4-year EXCEL data validation. It was subsequently replaced by the SYNTAX 2020 version, which demonstrated better adjustment in the validation of cohorts from the FREEDOM, PRECOMBAT, BEST, and a Japanese study for 10-year mortality and 5-year major cardiovascular events. The workgroup, however, does not endorse these systems, considering them underdeveloped. Despite previous calls to discontinue using the SYNTAX score as a decision-making axis, they continue to support its use, viewing it as the best available tool. It remains useful in determining whether coronary anatomy complexity exceeds 32 points—a key threshold in this guide.

Concerning the four analyzed studies, they update the latest follow-up results. Thus, the 5-year NOBLE study reported PCI vs. CABG mortality at 5% vs. 5%, p>0.05; new AMI at 8% vs. 3%, p <0.05; and revascularization need at 17% vs. 10%, p <0.05. In the 5-year EXCEL study, mortality was 13% vs. 9.9%, p <0.05; ischemia-driven revascularization 16.9% vs. 10%, p <0.05; and AMI, both periprocedural and new-onset according to the fourth international AMI definition, 9.6 vs. 4.7%, p <0.05. The 10-year PRECOMBAT study showed mortality at 14.5% vs. 13.8%, p >0.05; and new revascularization at 16.1% vs. 8%, p <0.05. In the 10-year SYNTAX study, mortality was 27% vs. 28%, p >0.05.

Using these data, the authors performed a new patient-level meta-analysis with follow-ups to the latest available dates, enabling results beyond 5 years. Including 4394 patients, 5-year PCI vs. CABG mortality was 11.2% vs. 10.2%, p =0.33, and beyond 10 years, 22.4% vs. 20.4%, p =0.25. Bayesian analysis favored surgery, albeit with a small margin. Other meta-analysis results included new AMI at 6.2% vs. 2.6%, p <0.0001; periprocedural AMI per study protocol (46% of the NOBLE, EXCEL, and complete SYNTAX samples) at 3.2% vs. 4.7%, p =0.13; and by the fourth AMI definition, 3.2% vs. 2.3%, p =0.15. CABG was superior for revascularization need at 18.3% vs. 10.7%, p <0.0001.

The guideline’s major conclusion is that in stable or stabilized LMCA disease following an acute event, with complexity according to SYNTAX score <33 and low-risk patients, surgery remains Class I, while PCI is downgraded to IIa (Class III for scores ≥33).

COMMENTARY:

As previously mentioned, maintaining the strength of the surgical indication and downgrading the percutaneous alternative, for a SYNTAX score <22 points to Class IIa, is again a milestone that looks promising for our interests but may not be sufficient in the future.

The authors conduct a parallel analysis of the existing gaps in knowledge in this field and the dependence that still exists on these now illustrious four studies. Among the critical comments made, several methodological warnings stand out, aimed at standardizing the way results are communicated in this and other areas of cardio-surgical controversy, which we list below and have previously referenced in earlier blog posts. First, the mortality considered acceptable is only all-cause mortality, as isolating cardiovascular mortality is misleading. This is because, precisely, some patient deaths from complications in other systems may stem from poor cardiovascular function, which would be artificially analyzed. Second, AMI should be defined based on the criteria of the fourth international definition, as a common framework for defining variables across all studies. Third, composite outcomes should be avoided, especially when one of them disproportionately influences and skews the overall result. This means that the primary outcome of any study cannot be a composite event, and each partial result that would make up the composite should be presented individually. Third, the analysis of stroke incidence is inconsistent across studies and should be the subject of in-depth analysis. This disparity occurs because, logically, the rates post-PCI are initially higher than those after surgery in the first year. However, during subsequent follow-up, an equalization occurs, the cause of which is poorly understood. Fourth, the analyses conducted in the studies do not consider outcomes that are highly significant in modern healthcare, such as life-years gained and their adjustment by both quality (utility) and value (considering the patient’s perception). It remains unknown whether the sequelae of an aggressive approach like surgery could have a negative impact or, conversely, whether the greater frequency of hospitalizations, symptom recurrence, and reinfarctions could burden the percutaneous alternative. This aspect is particularly relevant given the apparent long-term overall survival tie analyzed crudely, as many of the events experienced by patients following index revascularization are mostly non-fatal.

Within the classes of recommendations issued, the authors maintain a call for dialogue and continued teamwork within Heart Teams, with collegial and consensual decision-making involving the patient. They insist on keeping the SYNTAX score as the decision-making axis and weighing different aspects that lead to favoring one therapeutic option over the other. To this effect, they propose a table indicating that surgery will preferably be assigned in patients with: left ventricular ejection fraction <35%, diabetes mellitus, contraindications for dual antiplatelet therapy, failure of previous percutaneous revascularizations, distal or bifurcation LMCA lesions, multivessel disease, predictable incomplete percutaneous revascularization, occluded right coronary artery with graftable distal bed, severe calcification limiting adequate stent expansion, need for concomitant procedures. PCI would be preferable in patients with advanced age and/or low life expectancy, high morbidity, high surgical risk, prior CABG with a patent left internal mammary artery graft to the left anterior descending artery (protected LMCA), ostial or body LMCA lesions, and the classic inherited cases from the TAVI spectrum, such as previous chest radiotherapy, extreme chest deformity, and porcelain ascending aorta (although for the latter, they recognize possible adaptations of surgical technique like no-touch aortic surgery: without CPB and without proximal anastomoses donors in the ascending aorta with various graft configurations).

Lastly, the authors warn of the need to “retire” these four illustrious studies and call for the generation of new evidence, better adapted to our times. There are notable differences from current practice, such as the need to update the morbidity of the populations included in the studies (worse lifestyle habits and aging population), technical and care advances incorporated in both fields (use of multiple arterial grafts, advances in surgical and perfusion techniques, enhanced postoperative recovery protocols, IVUS, functional analysis of coronary lesions, new-generation drug-eluting stents that have left the paclitaxel-coated TAXUS from the SYNTAX study far behind), pharmacological advances (new statins and lipid-lowering therapies, new antiplatelets and oral anticoagulants, and the emergence of SGLT2 inhibitors), and a result analysis more aligned with modern healthcare objectives (efficiency, utility, and value).

With all this, coronary artery bypass surgery continues its steady course. However, after more than half a century of evolution, new milestones will emerge on the long road of revascularization history. This is merely the prelude to future clinical guidelines, long-awaited as they may be, or new studies that will challenge the paradigm we take for granted today.

REFERENCE:
Byrne RA, Fremes S, Capodanno D, Czerny M, Doenst T, Emberson JR, et al. 2022 Joint ESC/EACTS review of the 2018 guideline recommendations on the revascularization of left main coronary artery disease in patients at low surgical risk and anatomy suitable for PCI or CABG. Eur J Cardiothorac Surg. 2023 Aug 1;64(2). doi:10.1093/ejcts/ezad286.