This study evaluated the utility of the SYNTAX score for predicting cardiovascular events in patients with diabetes mellitus (DM) and complex coronary artery disease (CAD) undergoing coronary artery bypass grafting (CABG) or percutaneous coronary intervention (PCI) based on data from the FREEDOM trial. The study found that the SYNTAX score had a modest correlation with hard cardiovascular events and a significant correlation with major adverse cardiac and cerebrovascular events in the PCI group. However, the rate of major adverse cardiac and cerebrovascular events was higher after PCI compared to CABG across all SYNTAX score categories. Therefore, the SYNTAX score should not be used to guide the choice of revascularization for patients with DM and mult

1. SYNTAX Score in Patients With Diabetes
Undergoing Coronary Revascularization in
the FREEDOM Trial
JACC Vol 72, December 11 2018 Esper et al.
Presented by:
Dr. Vaibhav Yawalkar
MD, DM Cardiology

2. BACKGROUND
Diabetes mellitus is associated with complex coronary artery
disease, which in turn results in increased morbidity and
mortality from cardiovascular disease.
OBJECTIVES
This study sought to evaluate the utility of SYNTAX score for
predicting future cardiovascular events in patients with DM
and complex CAD undergoing either coronary artery bypass
grafting (CABG) or percutaneous coronary intervention (PCI).

3. METHODS
The FREEDOM (Future REvascularization Evaluation in
patients with Diabetes mellitus: Optimal management of
Multivessel disease) trial randomized patients with DM and
multivessel CAD to undergo either PCI with drug-eluting stents
or CABG.
The Syntax Score was calculated retrospectively by a core
laboratory.

4. The endpoint of hard cardiovascular events (HCE) was a
composite of
• death from any cause
• nonfatal myocardial infarction
• nonfatal stroke
The endpoint of major adverse cardiac and cerebrovascular
events (MACCE) was a composite of
• HCE (as mentioned above)
• Repeat revascularization.

5. INTRODUCTION
In patients with diabetes mellitus, compared with those
without DM, CAD tends to be more diffuse, complex, and
associated with increased morbidity and mortality.
For patients with DM and multivessel CAD, coronary artery
bypass grafting (CABG) has been definitively shown to
improve outcome compared with percutaneous coronary
intervention (PCI).
Specifically, the FREEDOM trial showed a reduction in
mortality and in the primary composite outcomes of death,
myocardial infarction (MI), or stroke following CABG
compared with PCI

6. Whether based on lesser lesion complexity can we identify
a group of diabetic multivessel CAD patients, who will have
results following PCI, comparable to those following CABG
is not known.
Hence there is not enough data for preference to PCI over
CABG in such group.

7. SYNTAX score
The SYNTAX score uses coronary anatomy to objectively
guide decisions regarding mode of revascularization.
Studies have confirmed the clinical validity of the Syntax
Score for identifying higher-risk subjects and aiding decision
making between CABG and PCI in a broad range of patient
types.
The Syntax Score is now advocated in both the European
and U.S. revascularization guidelines as an aid for deciding
on revascularization strategy.

8. SYNTAX score
The results of the SYNTAX (Synergy between PCI with Taxus
and Cardiac Surgery) trial suggest that,
The Syntax Score not only is a valid tool for estimating the
risk associated with PCI in patients with left main or
multivessel CAD, but also may be an effective aid in selecting
those with outcomes after PCI that would be predicted to be
similar to CABG

10. SYNTAX score II (Anatomical+ Clinical)
 SYNTAX score I
 Age
 Creatinine Clearance
 LVEF
 Left main to bifurcation lesion
 Gender
 COPD
 PVD

11. METHODS
This study is a post hoc analysis using the subjects enrolled
in the FREEDOM trial.
The FREEDOM trial was a multicenter trial that studied DM
patients with multivessel CAD without LMCA stenosis or
chronic total occlusions that had an indication for
revascularization & were candidates for either CABG or PCI.
Patients were randomized to either CABG or PCI with drug-
eluting stents.
Multivessel CAD was defined as stenosis of > 70% in 2 or
more major epicardial vessels involving at least 2 separate
coronary artery territories.

12. REVASCULARIZATION AND PHARMACOLOGIC THERAPY
First-generation sirolimus-eluting stent
Cypher stent
Cordis
Johnson & Johnson, Miami Lakes, Florida
Paclitaxel eluting stent
Taxus stent
Boston Scientific ,Boston, Massachusetts
The interventional cardiologist was encouraged to treat all
arteries that were likely to contribute to ischemia or had lesions
with >70% diameter stenosis and to achieve complete anatomic
revascularization.

13. The use of glycoprotein IIb/IIIa inhibitors was recommended
for patients undergoing PCI.
Dual antiplatelet therapy with aspirin and clopidogrel was
recommended for at least 12 months after stent implantation
along with antianginal therapy.
Use of internal mammary conduits was strongly advised for
all CABG cases.
The recommended target levels:
low-density lipoprotein cholesterol <70 mg/dl
blood pressure < 130/80 mm Hg
glycated hemoglobin < 7%

14. Syntax Score CALCULATION
The Syntax Score for each patient was calculated
retrospectively by scoring all coronary lesions with a
diameter stenosis > 50% in vessels > 1.5 mm, using the
Syntax Score algorithm.
The Syntax Score for each patient was calculated
retrospectively by the core laboratory at the Cardiovascular
Research Foundation.
The subjects were categorized according to the level of the
Syntax Score:
Low : 0 to 22
Intermediate 23 to 32
High > 33

15. RESULTS
Duration: From April 2005 to April 2010
32,966 patients were screened to FREEDOM trial.
Of the 3309 trial-eligible patients, 1900 (57.4%) provided
written informed consent and underwent randomization.
A total of 1900 patients,
953 in the PCI group and
947 in the CABG group,
were followed prospectively.
The overall clinical, laboratory, and angiographic
characteristics were well balanced in the 2 groups and for
each Syntax Score category

16. The median overall follow-up for all patients was 3.8
(interquartile range 2.5 to 4.9) years.
The median follow-up until the first MACCE event (or
last-seen date) was 3.2 (IQR: 1.9 to 4.5) years
The median follow-up until and the first HCE event (or
last-seen date) was 3.5 (IQR: 2.3 to 4.6) years.

17. PCI group, Events as per Syntax Score Categories

18. CABG group, Events as per Syntax Score Categories

19. PCI group, Event Predictors at 5 years follow up

20. MACCE PCI Group HCE PCI Group
Although in PCI group the Syntax Score was an independent risk factor for
MACCE and HCE at 5 years, the area under the ROC curve showed a poor
discrimination capability for MACCE (0.54) and HCE (0.56)

21. MACCE CABG Group HCE CABG Group

22. CABG group, Event Predictors at 5 years follow up
The Syntax Score was not an independent risk factor for MACCE and HCE in
the CABG group

23. Kaplan-Meier Survival Curves of MACCE

25. Kaplan-Meier Survival Curves of HCE

27. SYNTAX Score vs Events

28. This study demonstrates that the Syntax Score has a
modest correlation with HCE and a significant correlation
with MACCE in patients with DM and multivessel disease
undergoing PCI.
Additionally, the MACCE rate was higher after PCI compared
with CABG for every tercile of Syntax Score, thus
demonstrating that the Syntax Score does not identify a
population of DM patients with multivessel disease in whom
PCI is equivalent or superior to CABG
Discussion

29. The BARI (Bypass Angioplasty Revascularization
Investigation) trial compared CABG with balloon angioplasty
and showed better survival in patients with diabetes who
received CABG rather than balloon angioplasty
The FREEDOM trial provided definitive proof of the
superiority of CABG compared with PCI in multivessel CAD
patients with diabetes.
Current U.S. guidelines recommend CABG as the preferred
coronary revascularization procedure in diabetic patients
with multivessel CAD
Previous studies

30. Syntax Score should be calculated for 2 potential major
reasons:
1) To predict the risk of future clinical events
2) To guide clinical decision making regarding the choice of
coronary revascularization strategy
Some studies have shown that angioplasty may be an
alternative to CABG in patients with less complex CAD (low
Syntax Score), particularly with current-generation drug-
eluting stents as recently reported in the SYNTAX 2 study.
In present study however, the incidence of MACCE was
higher in the PCI group regardless of Syntax Score. Unlike
findings from the SYNTAX trial, the FREEDOM trial
demonstrate that Syntax Score should not guide decision
making in diabetic patients with multivessel disease.

31. Inthis study, the Syntax Score was not an independent risk
factor for HCE and MACCE in the CABG group, a result
that is similar to other studies.
The Syntax Score includes anatomic variables such as
calcification, tortuosity, type of bifurcation, and extent of
CAD. However, for the patients undergoing CABG, a major
determinant of graft patency is a satisfactory anastomosis
to a suitable distal portion of the bypassed artery.
So, despite the complexity of CAD in proximal arterial
segments (which determine the Syntax Score), late post-
CABG outcomes are largely influenced by distal vessel
suitability for grafting, which generally makes minor
contributions to the Syntax Score.

32. Several studies have suggested that the Syntax Score
carries a significant prognostic role, with higher Syntax
Score associated with higher burden of long-term mortality
and ischemic adverse events in patients undergoing PCI.
Most of these studies include subjects with left main
disease, which were excluded from the FREEDOM trial.
Anatomic data alone, particularly when associated with left
ventricular function, is also a strong prognostic predictor,
comparable to the Syntax Score

33. Many reasons could explain why the Syntax Score has
modest discrimination capability for hard cardiovascular
events in diabetic patients.
The score considers only angiographic variables and does
not include clinical variables.
The diameter stenosis severity, unless the artery is totally
occluded (100%), does not affect the Syntax Score, with
coronary stenoses of 50%, 70%, or 95% stenosis scoring the
same value.
The score does not consider myocardial ischemia or
functional information. Coronary lesions in small arteries
between 1.5 and 2.0 mm in diameter are included in the
Syntax Score, but most of these lesions will not undergo
revascularization

34. This study shows that among diabetic patients, clinical
characteristics such as LVEF < 40%, insulin-dependent DM,
and chronic renal dysfunction are important predictors of
cardiovascular events and should be incorporated into the risk
prediction for coronary revascularization.
Scores that include clinical and angiographic variables, such
as Syntax Score II , clinical Syntax Score, and Global Risk
Score (EuroSCORE [European System for Cardiac Operative
Risk Evaluation] associated with Syntax Score) could be more
suitable to predict MACCE and HCE in diabetic patients.

35. STUDY LIMITATIONS
This study being ad hoc analysis, is not powered to make
comparisons between Syntax Score subgroups overall or
for any individual component of MACCE; therefore, these
results should be regarded as observational and
hypothesis-generating only and need to be confirmed in
subsequent adequately powered clinical trials.

36. SUMMARY
In DM with multivessel CAD, the incidence of MACCE is
higher for the PCI compared with CABG in all Syntax Score
categories.
The Syntax Score should not be utilized to guide the choice
of coronary revascularization in patients with DM and
multivessel CAD

37. Editorial Comment
Based on accumulated evidence, most experts strongly
advocate CABG as the preferred mode of revascularization
for people with diabetes and MVD.
Whether subsets of MVD patients with diabetes, with a
lower severity of coronary disease burden, may have
equivalent outcomes with PCI compared with surgery has
not been studied.
This new analysis shows that the SYNTAX score can
indeed predict outcomes in people with diabetes and MVD
undergoing PCI. Importantly, the major adverse cardiac and
cerebrovascular event rate was worse with PCI compared
with CABG, even in patients with the lowest SYNTAX
scores (and least burden of disease).

38. Editorial Comment
This study confirms that the SYNTAX score cannot be used
to identify selected patients with diabetes and MVD in
whom PCI may be superior to surgical revascularization.
These findings are at odds with prior efforts to answer the
same important question. Cavalcante et al. performed a
comparable analysis using a pooled dataset from 3 large
randomized controlled trials (SYNTAX, PRECOMBAT &
BEST) in patients with diabetes and MVD.
Their study showed a significant interaction between
SYNTAX score and mode of revascularization. The data
indicated that PCI and CABG have comparable 5-year
outcomes for all SYNTAX scores except for the highest
tercile, where CABG was superior.

39. These findings are supported by a recent meta-analysis of
pooled data from both the FREEDOM and SYNTAX studies
performed by Hakeem et al, who compared CABG and PCI
for major adverse cardiac events.
But compared with the FREEDOM study design, the
SYNTAX, PRECOMBAT, and BEST studies used a core
laboratory to more objectively determine SYNTAX scores,
which may increase the reproducibility of their findings
However, these studies also had much lower rates of left
internal mammary artery to left anterior descending artery
grafting & had suboptimal post-CABG medical therapy.
Thus, the comparison may have been insensitive to
potential differences due to suboptimal CABG outcomes.
Editorial Comment

40. The optimal revascularization strategy for diabetic patients
with MVD is CABG, and it should always be entertained as
the preferred strategy. However, we must tailor our decisions
by integrating available clinical factors such as operative
risks (especially stroke), predicted longevity, and patient
preferences for a physical recovery profile.
The recent 2018 ESC/EACTS guidelines on myocardial
revascularization recommend CABG for people with diabetes
and MVD irrespective of the SYNTAX score
Guidelines also recognize the appropriateness of PCI in
such patients when the surgical risk is elevated or when the
patient refuses surgical approaches
Editorial Comment

41. Mancini et al. have rigorously evaluated the effects of angina
severity, coronary disease burden, and left ventricular
function in patients with DM & MVD
The investigators simultaneously evaluated the influence of
optimal medical therapy, in addition to mode of
revascularization, on important clinical outcomes. The results
confirm that surgical revascularization provides the best
outcomes for people with diabetes and MVD.
Coronary bypass with optimal medical therapy was superior
over PCI irrespective of angina severity, disease burden, or
baseline left ventricular function. With a high SYNTAX score,
patients should be strongly encouraged to have surgical
revascularization if suitable candidates.
Editorial Comment

42. Interestingly, outcomes with optimal medical therapy
alone (no revascularization) were similar to PCI.
Accordingly, when the SYNTAX score is low, optimal
medical treatment should be entertained (compared with
PCI) if coronary bypass is not feasible or acceptable to the
patient.
If optimal medical therapy fails to control symptoms, only
then should PCI be considered.
Editorial Comment

43. We can safely conclude that all suitable patients with
diabetes and MVD with indications for revascularization
should be strongly considered for CABG as the preferred
revascularization strategy.
The SYNTAX score should not be applied in this specific
decision making process.
Clinical characteristics such as LVEF < 40%, insulin-
dependent DM, and chronic renal dysfunction are
important predictors of cardiovascular events and should
be incorporated into the risk prediction for coronary
revascularization.
Take Home Message