| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a Department of Cardio-Thoracic Surgery, Erasmus MC Rotterdam, the Netherlands
b Department of Cardiology, Erasmus MC Rotterdam, the Netherlands
c Department of Cardiac Surgery, Heart Center, University of Leipzig, Germany
Accepted for publication November 10, 2008.
* Address correspondence to Dr Birim, Department of Cardiothoracic Surgery, Room BD 575, Erasmus MC, P.O. Box 2040, Rotterdam, 3000 CA, the Netherlands (Email: o.birim{at}erasmusmc.nl).
Presented at the Forty-fifth Annual Meeting of The Society of Thoracic Surgeons, San Francisco, CA, Jan 26–28, 2009.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterial and MethodsResultsCommentDiscussionReferences |
|---|
Methods: One hundred forty-eight patients were studied. Their angiograms were scored according to the SYNTAX score. The MACCE-free survival curves were estimated by the Kaplan–Meier method. Univariate and multivariate analyses determined risk factors for MACCE. Performance of the SYNTAX score was studied with respect to discrimination by receiver-operating characteristic curves with their area under the curve (c-index). Classification and regression tree analysis was performed to identify the best outcome predictors and develop a risk stratification model.
Results: Overall SYNTAX score ranged from 11 to 53 (mean, 24 ± 9). At 30 days and 1 year, 15 (10%) and 19 (13%) patients experienced MACCE. Patients with a higher SYNTAX score had a significantly (p < 0.0001) poorer MACCE-free survival. In multivariate analysis, SYNTAX score, female sex, and incomplete revascularization were associated with a higher rate of MACCE in 30 days. The SYNTAX score was the single predictor for MACCE in 1 year. The c-index of the SYNTAX score was 0.88 for 30 days and 0.90 for 1 year, respectively. The SYNTAX score was the best single discriminator between patients with and those without MACCE, with a discrimination level of 36.5.
Conclusions: The SYNTAX score is the first coronary vasculature complexity score predictive for postoperative outcome in patients with left main coronary artery disease undergoing coronary artery bypass grafting. The outcomes of the ongoing SYNTAX trial will definitively define the role of the SYNTAX score in predicting short-term and long-term incidence of MACCE.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionReferences |
|---|
|
Material and Methods |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionReferences |
|---|
SYNTAX Score and Angiographic Analysis
Each coronary lesion producing 50% or greater luminal obstruction in vessels 1.5 mm or greater was separately scored and added to provide the overall SYNTAX score. The SYNTAX score was calculated using dedicated software that integrates the number of lesions with their specific weighting factors based on the amount of myocardium distal to the lesion according to the score of Leaman and colleagues [15] and the morphologic features of each single lesion, as previously reported [11]. An example of the SYNTAX score calculation in 1 patient is shown in Figure 1. The patient's SYNTAX score was stratified into SYNTAX score tertiles [12]. All diagnostic angiograms were scored by one experienced investigator (Ö.B.) who was blinded as to procedural data and clinical outcome.
|
End Point Definitions
Death from all causes was reported. A cerebrovascular accident is any acute event related to the impairment of the cerebral circulation that lasts more than 24 hours and results in irreversible brain damage or permanent body impairment. Myocardial infarction was considered if there was documentation of new abnormal Q waves and a ratio of serum creatinine kinase-MB isoenzyme to total cardiac enzyme that was greater than 0.1 or a creatinine kinase to creatinine kinase-MB value that was five times the upper limit of normal [10]. Serum creatinine kinase and creatinine kinase-MB isoenzyme concentration were measured 6, 12, and 18 hours after operation. All repeat revascularization procedures by either PCI or CABG were recorded. Events were counted from the time of operation.
Statistical Analysis
Discrete variables are displayed as proportions, continuous variables as mean ± standard deviation unless specified otherwise. The 
2 (whenever n > 5 in all groups) or Fisher's exact test was used to analyze the categorical data. Differences between continuous variables were analyzed using one-way analysis of variance. When comparing three groups, a probability value of less than 0.0167 was considered significant (
correction according to Bonferroni). The MACCE-free survival curves were estimated by the Kaplan–Meier method. Differences in survival were compared using the log-rank test. Univariate and multivariate logistic regression analysis determined risk factors for MACCE within 30 days. Univariate and multivariate Cox proportional hazard analysis determined risk factors for MACCE within 1 year. A probability value of less than 0.05 was considered significant. The multivariate analyses were performed with a stepwise backward regression model in which each variable with a probability value of less than 0.20 in the univariate analysis was entered in the model. Relative risks are reported with 95% confidence intervals. Performance of the SYNTAX score was studied with respect to discrimination (resolution). Discrimination refers to the ability to distinguish patients with MACCE from those without. It was assessed by receiver-operating characteristic curves with their areas under the curve (c-index) with 95% confidence limits. A c-index of 1.0 would indicate perfect discrimination, whereas a c-index of 0.5 indicates total absence of discriminative power.
All variables associated with the incidence of MACCE at 1 year at a probability value of 0.10 in the Cox proportional hazard analysis were subjected to classification and regression tree analysis to identify the best outcome predictors and develop the risk stratification model [16]. This method is based on recursive partitioning analysis and involves the segregation of different values of classification variables through a decision tree composed of progressive binary splits. This approach has the advantage of uncovering possible interactions among predictors.
Descriptive statistical analyses were performed with SPSS 15.0 for Windows (SPSS, Chicago, IL), and R version 2.5.1 (R Foundation for Statistical Computing, Vienna, Austria) was used for calculating c-indices with 95% confidence limits, constructing receiver-operating characteristic curves, plotting Kaplan–Meier survival curves, and performing classification and regression tree analysis.
|
Results |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionReferences |
|---|
|
|
|
|
|
|
By evaluating all variables related to 1-year MACCE at a p value of 0.10 or less at univariate Cox proportional hazard analysis (Table 4), the classification and regression tree method confirmed the SYNTAX score as the best single discriminator between patients with and those without MACCE, with a discrimination level of 36.5 (Fig 4). When stratified into the discrimination level suggested by classification and regression tree analysis, the adjusted relative risk for MACCE was 21.5 (95% confidence interval, 8.3 to 55.9; p < 0.0001) for patients with a high versus low SYNTAX score.
|
|
Comment |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionReferences |
|---|
Most prognostic models proposed thus far have been derived from an original dataset from a large-scale registry or a randomized controlled trial [17]. In this context, a vital aspect of prediction is to consider whether such a model is applicable to similar patients in another setting. A model that is found to pass such a test is said to have been validated [18]. The SYNTAX score was created by an international group of expert interventional cardiologists and cardiac surgeons by merging together and tailoring several previously proposed coronary artery disease scoring systems based on personal expertise [11]. The ultimate goal is to create an angiographic tool grading the complexity of coronary artery disease and obtain evidence-based guidelines for selecting the optimal technique of revascularization (CABG or PCI). The present report is the first evaluation of the predictive value of this recently developed angiographic scoring system in patients undergoing CABG. However, the SYNTAX score cannot be considered fully validated because this is the only data set of patients undergoing CABG in which the model has been tested. As such, it remains unclear whether and to what extent our present findings can be reproduced in a different group of patients with left main coronary artery disease. The outcomes of the SYNTAX trial, which is ongoing at the present time, in which the SYNTAX score will be used to predict clinical outcomes at 1 month, and 1, 3, and 5 year after the procedure will most likely define the role of the SYNTAX score in predicting clinical outcomes after CABG or PCI [10, 11].
Interpretation of diagnostic angiograms, as any clinical tool, is subject to some intraobserver and interobserver variability. A limitation of the present study is that one experienced investigator scored the angiograms, whereas in a prospective randomized controlled trial the angiograms are scored by a local heart team (composed of both a cardiothoracic surgeon and an interventional cardiologist), and as a consequence probably will decrease the interobserver variability. The present study is performed in a relatively small number of patients. Therefore, some covariates, such as age and urgent surgery, which have been shown to affect outcome in earlier studies, might have been excluded from the multivariate model.
In conclusion, the SYNTAX score is the first coronary vasculature complexity score predictive for postoperative outcome in patients with left main coronary artery disease (isolated, or in combination with one-, two-, or three-vessel disease) undergoing CABG. The outcomes of the ongoing SYNTAX trial [19] will definitively define the role of the SYNTAX score in predicting short-term and long-term incidence of MACCE.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionReferences |
|---|
And the second question is, have you had fractional flow reserve as an additional tool to discriminate those who will have a higher MACCE?
DR BIRIM: Thank you for your questions.
To answer question one first, we only analyzed patients with left main disease not patients with only 3-vessel disease. But maybe the outcomes of the SYNTAX trial will answer this question properly and define which patients with 3-vessel disease are at higher risk than other patients with 3-vessel disease.
To answer you second question, no, we didn't do an FFR reading within these patients because it was a retrospective study and we only checked the angiograms and scored their SYNTAX score.
DR L. LESTER (West Palm Beach, FL): It seems remarkable that here we are into the 50th year of coronary bypass and we still haven't really categorized anything beyond one, two and three. And yet we all know that operating on patients with extremely diffuse disease is a significant risk. It's really kind of a remarkable phenomenon.
Dalhousie University in Nova Scotia has looked at the risk model in terms of whether it adversely affects patients who have renal failure, whether it's the diffuse disease or the impact of renal failure. But in their series of patients, those who had, by their scoring system, greater than 18, had an operative mortality of 8% or 9%, and those who had a score of less than 18 had only a 2%. And certainly there is a large population of patients that we consider inoperable because their mortalities are substantially higher than our typical patients who can benefit from revascularization. And historically, I've done a fair number of patients with extremely diffuse disease and their mortality was about 10%, which is what Dalhousie showed, but they really do better than they do medically because they're usually in terrible jeopardy with very diffuse disease.
And I just can't imagine that as a Society that we don't have some emphasis, at least in patients designated by the surgeon, to score them so that we can adequately risk model them. It's almost as if plumbers were not allowed to describe the plumbing in your home when they came to work there. It's bizarre, actually, that we do it this way. I applaud this effort to define a central and neglected issue in risk model of coronary patients.
DR BIRIM: I agree.
DR D. CHU (Houston, TX): It's a great paper. I have two questions.
Do you think that as part of the SYNTAX score the quality of the coronary target should be a metric as part of the SYNTAX score? Because we all know that the patient could have multiple bad calcifications and multiple segmental disease. If they have adequate distal target, you pull them through with a CABG without any problems. So I don't notice the quality of distal target as part of the SYNTAX score, number one.
Number two, we all know that incomplete revascularization increases MACCE and decreases long-term survival in patients with CAD. And I noticed that in your patients, there is a subgroup where the SYNTAX was greater than 25, yet only 78% of them achieved complete revascularization; that is, the other 21% didn't. Do do you think that the SYNTAX score is simply a surrogate marker for incomplete revascularization or vice versa?
DR BIRIM: Thank you for your questions.
To answer question one first, the SYNTAX score is based on multiple scoring systems and multiple factors influence the SYNTAX score. So a couple of these factors are already included in the score, like degree of calcification, which also includes the distal part of a lesion, and total occlusions. If you have total occlusions on the angiogram, you don't know what the distal vessel looks like. It can be a good vessel, it can be a bad vessel. So these kinds of things are already incorporated in the SYNTAX score.
And to answer your second question, we looked at the incomplete revascularization rate at our univariate and multivariate analysis, and the incomplete revascularization was only a factor in univariate analysis and not in multivariate analysis. Maybe it's because our patient population wasn't high enough, it was only 148 patients. But I think incomplete revascularization is indeed a factor influencing survival in MACCE postoperatively. However, when you look at the angiogram and you look at the SYNTAX score, you do it preoperatively and you don't know how complete or incomplete your revascularization is.
DR F. W. MOHR (Leipzig, Germany): I think you made a very nice presentation. And we, as coauthors, face a real problem here because what you tell us is not what we tell you. And just if you have listened to Mike Mack yesterday, and maybe at the EACTS when Patrick and I presented the data about the SYNTAX trial, they are absolutely contradictory to your data. If you take the 1,800 patients, being operated upon during the SYNTAX trial and the classification with the SYNTAX score was performed it was quite obvious that in the randomized cohort, the SYNTAX score was lower in total left main and 3-vessel disease, or in the combination of both.
Whereas in the CABG registry the SYNTAX score was around 34 in yet the MACCE rate was only 8.8 at one year compared to 12.1 in the randomized cohort which had lower SYNTAX scores. So what that means, there is something in your data which is hardly understandable from the perspective of the SYNTAX trial because your so-called high score patients (25–30) the mortality was up to 16%.
My question is whether the numbers, the 194 patients you looked at, may be somewhat misleading, and we should not, with all respect, overestimate these conclusions.
Because if you look at real numbers in the SYNTAX trial, the SYNTAX score provides a description of the complexity of the coronary disease, but it affected more the interventional cardiologists by PCI rather than the surgeons. Actually it's almost the opposite, the higher the score, the more total occlusions you see, the better is the surgical outcome. We do not fully understand that right now. Maybe it's less competitive flow.
In this respect your paper is somehow disturbing for me.
DR BIRIM: If I may comment on this question.
First of all, there are a couple of limitations in our study. It's a retrospective study with only 148 patients included. And our angiograms were scored by one investigator only.
Second of all, I was surprised by this study also because there are a lot of factors in the SYNTAX score that only affect PCL, you would say, like tortuosity of the vessels, total occlusion of the vessels, and degree of calcification at the lesion location itself.
DR MOHR: But it's also calcification, diffuse calcification and disease, you would also account for that in the SYNTAX score.
DR BIRIM: So that was the purpose of our study, to validate the score in patients undergoing CABG, because we also thought that it was maybe more appropriate to score for PCI patients than for CABG patients.
On the other hand, we only looked at left main stenosis patients. And it could be that the difference in SYNTAX score and degree of complexity of the coronary vessels is more obvious in patients with left main disease than in patients with 3-vessel disease. And I noticed that in the SYNTAX trial only one-third of patients have left main disease and two-thirds of patients have 3-vessel disease. So it can be that it is more obvious in left main disease patients than in 3-vessel disease patients.
DR MOHR: But left main disease alone has very low scores such as a score of 5, and we have observed a MACCE of 12% in that study. So this is even worse if you take patients who have a real high complex lesion set.
DR A. P. KAPPETEIN (Rotterdam, the Netherlands): Fred, you're right, these data are in contradiction with the results of the SYNTAX trial, In the SYNTAX trial a higher SYNTAX score is not predictive for MACCE in the CABG cohort while it is predictive for MACCE in PCI patients. I think, however, that we cannot compare the group of patients presented here with the randomized patients of the SYNTAX trial. The patients presented in the current cohort have also been discussed in the heart team and the cardiologists denied PCI. These patients are therefore comparable with the registry patients of the SYNTAX trial.
DR MOHR: But this was 2001 to 2004, so this was a little bit before the real aggressive area of the interventional cardiologists.
DR KAPPETEIN: However, left main and three vessel disease are already treated for a long time in our center. In my opinion you can compare these patients better with the registry patients of SYNTAX. Also in this registry cohort of SYNTAX, MACCE difference between the lower and the higher SYNTAX score, is small. However, we find a cutoff point, in this study it was 36, it might be that in the SYNTAX trial we will find a cutoff point of, let's say, 40 or 45. But the more complex the coronary artery disease is, the higher the risk for patients. Peripheral vascular disease, was also more prevalent in the patients with a higher SYNTAX score and indicates the extent of vascular disease. It might be a question of finding the right cutoff point.
|
References |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  M. Carnero-Alcazar, L. C. Maroto Castellanos, J. A. Silva Guisasola, J. C. Carnicer, A. Alswies, M. E. Fuentes Ferrer, and J. E. Rodriguez Hernandez SYNTAX Score is associated with worse outcomes after off-pump coronary artery bypass grafting surgery for three-vessel or left main complex coronary disease J. Thorac. Cardiovasc. Surg., September 1, 2011; 142(3): e123 - e132. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
F. W. Mohr, A. J. Rastan, P. W. Serruys, A. P. Kappetein, D. R. Holmes, J. L. Pomar, S. Westaby, K. Leadley, K. D. Dawkins, and M. J. Mack Complex coronary anatomy in coronary artery bypass graft surgery: Impact of complex coronary anatomy in modern bypass surgery? Lessons learned from the SYNTAX trial after two years. J. Thorac. Cardiovasc. Surg., January 1, 2011; 141(1): 130 - 140. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Chikwe, M. Kim, A. B. Goldstone, A. Fallahi, and T. Athanasiou Current diagnosis and management of left main coronary disease Eur J Cardiothorac Surg, October 1, 2010; 38(4): 420 - 430. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Garg, G. W. Stone, A.-P. Kappetein, J. F. Sabik III, C. Simonton, and P. W. Serruys Clinical and Angiographic Risk Assessment in Patients With Left Main Stem Lesions J. Am. Coll. Cardiol. Intv., September 1, 2010; 3(9): 891 - 901. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. Capodanno, P. Capranzano, M. E. Di Salvo, A. Caggegi, D. Tomasello, G. Cincotta, M. Miano, M. Patane, C. Tamburino, S. Tolaro, et al. Usefulness of SYNTAX Score to Select Patients With Left Main Coronary Artery Disease to Be Treated With Coronary Artery Bypass Graft J. Am. Coll. Cardiol. Intv., August 1, 2009; 2(8): 731 - 738. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |
