Prognostic value of clinical SYNTAX score on 2-year …SS is a scoring system based on the complexity and severity of coronary lesions. The calculation method takes coronary dominance

Prognostic value of clinical SYNTAX score

on 2-year outcomes in patients with ACS

undergoing PCI

Peking Union Medical College

Fuwai Hospital

Chen He; Na Xu; Jin-QingYuan;

Disclosure Statement of Financial

Interest

The authors declared no conflict of interest

Clinical SYNTAX Score combined the SS and a variant of

the ACEF (age, creatinine and left ventricular ejection

fraction) score, was thought to be a simple and convenient

prognostic tool to predict long-term outcomes.

Background

This prospective, single-center, observational study

evaluated prognostic value of clinical SYNTAX score on

2-year outcomes in patients with ACS undergoing PCI

Calculation of Clinical SYNTAX Score

• The CSS formula CSS=[SYNTAX

Score]*[modified ACEF score].

• The modified ACEF was calculated

retrospectively using the formula Age/EF+N

N=every 10mL/min reduction in CrCl below

60 mL/min /1.73 m2 (up to a maximum of 6

points).

Study Design

10724 consecutive patients with PCI in

Fuwai Hospital (2013.01-2013.12)

6099 patients were enrolled

low CSS group

CSS≤6.5

(2012 patients)

Inclusion criteria: ACS including

STEMI/NSTEMI/UAP

Exclusion criteria:

previous CABG, SAP

2 years follow-up Drug treatment was taken

following guideline

Complete clinical follow-up information for 2 years was available for

2002 patients in the low CSS group (99.5%), 2044 patients in the mid

CSS group (99.4%) and 2020 patients in the high CSS group (99.5%)

number of clinical

SYNTAX score

mid CSS group

6.5＜CSS＜13.8

(2056 patients)

high CSS group

CSS≥13.8

(2031 patients)

Result

Table 1: Baseline clinical characteristics of study patients Variable CSS≤6.5

(n=2012)

6.5＜CSS＜13.8

(n=2056)

CSS≥13.8

(n=2031)

p Value

Age (years ) 54.6± 9.7 57.1± 9.7 63.1± 10.1 <0.001

Men 1604(80%) 1621(79%) 1456(72%) <0.001

Hypertension 1224(61%) 1288(63%) 1344(66%) 0.002

Hyperlipidemia 1318(66%) 1389(67%) 1317(65%) 0.818

Diabetes Mellitus 502(25%) 582(28%) 668(33%) <0.001

Previous cerebrovascular accident 154(7.7%) 226(11.0%) 270(13.3%) <0.001

Current smoker 1233(61%) 1257(61%) 1071(60%) 0.621

Family history of CAD 522(26%) 531(26%) 453(25%) 0.578

Prior myocardial infarction 194(10%) 244(12%) 336(17%) <0.001

Prior PCI 410(20%) 417(20%) 480(24%) 0.013

Body Mass index (kg/m2) 26.1± 3.1 26.0± 3.2 25.5± 3.2 <0.001

Brain Natriuretic Peptide (ng/L) 646± 360 721± 444 973± 917 <0.001

GFR 96± 13 93± 13 84± 18 <0.001

Uric acid (µmol/L) 344± 83 341± 95 338± 89 0.110

Total Cholesterol (mg/dl) 163± 41 163± 40 163± 42 0.987

Low density lipoprotein (mg/dl) 97± 35 98± 34 98± 36 0.557

High density lipoprotein (mg/dl) 40± 11 39± 11 40± 11 0.339

Triglyceride (mg/dl) 164± 105 160± 93 157± 89 0.12

Ejection fraction 64.4%± 5.9% 62.9%± 6.7% 59.8%± 8.5% <0.001

Baseline SYNTAX score 4.7± 2.2 10.8± 3.3 20.2± 6.9 <0.001

Table 2: Angiographic and procedural characteristics of patients

Variable CSS≤6.5

(n=2012)

6.5＜CSS＜13.8

(n=2056)

CSS≥13.8

(n=2031)

p Value

Left main stenosis 7(0.3%) 27(1.3%) 110(5.4%) <0.001

Single vessel disease 1812（90%） 1521(74%) 1312(65%) <0.001

STEMI 392(19.5%) 459(22.3%) 569(28.0%) <0.001

NSTEMI 135(6.7%) 152(7.4%) 171(8.4%) <0.001

Unstable angina 1485(74%) 1445(70%) 1291(64%) <0.001

IABP 4(0.2%) 14(0.7%) 79(3.9%) <0.001

Radial artery puncture 1898(94%) 1945(95%) 1828(90%) <0.001

Stent implantation 1946(97%) 1955(95%) 1887(93%) <0.001

Successful PCI 1989(99%) 1990(97%) 1882(93%) <0.001

No. of stents per patient 1.4± 0.7 1.8± 1.0 2.1± 1.3 <0.001

Any bare-metal stent 8(0.4%) 8(0.4%) 14(0.7%) 0.297

GFR after PCI (ml/min) 91± 14 88± 14 81± 16 <0.001

Time of PCI (min) 25± 16 36± 31 47± 37 <0.001

Residual SYNTAX Score 0.6± 1.4 2.6± 3.5 7.1± 7.9 <0.001

Hospital stay (days) 5.3± 2.6 6.0± 3.4 7.5± 4.9 <0.001

Medical treatment in hospital

Aspirin 1987(99%) 2024(98%) 1993(98%) 0.272

Thienopyridines 1982(99%) 2022(98%) 1986(98%) 0.188

Beta-blockers 1744(87%) 1857(90%) 1833(90%) <0.001

Statins 1938(96%) 1977(96%) 1930(95%) 0.08

Unfractionated heparin 1230(61%) 1248(61%) 1222(60%) 0.82

GP IIb/IIIa use 272(14%) 331(16%) 361(18%) 0.001

Long-term cardiac events of all study patients

0

2

4

6

8

10

12

14

16

18

low CSS group mid CSS group high CSS group

P<0.001

P=0.009 P=0.036

P=0.664

P<0.001

P<0.001

Kaplan-Meier plot for long-term mortality

Table 4: Independent predictors of cardiac death and MACE according to multivariate

Cox analyses

Event Variables HR(95%CI) p Value

Cardiac death

Clinical SYNTAX score 1.025(1.018-1.032) <0.001

Prior PCI 2.714(1.440-5.116) 0.002

Hypertension 2.526(1.114-5.730) <0.001

MACE Clinical SYNTAX score 1.008(1.005-1.012) <0.001

IABP support 2.232(1.520-3.279) <0.001

Diabetes mellitus 1.250(1.073-1.457) 0.004

Successful PCI 1.575(1.335-1.858) <0.001

SYNTAX

Score

SS is a scoring system based on the complexity and severity of

coronary lesions. The calculation method takes coronary dominance

variants into account and applies different segment weighting

factors in different coronary dominance, depending on the coronary

blood flow distribution

Serruys PW, Onuma Y, Garg S, et al. Assessment of the SYNTAX

score in the Syntax study. EuroIntervention. 2009. 5(1): 50-6.

Improve the accuracy and specificity of

SYNTAX Score

SYNTAX Score

Clinical SYNTAX Score

Logistic CSS Residual

SYNTAX Score

SYNTAX II

ROC curves analysis for cardiac death

CSS improved the

discriminatory power

compared with SS in

predict 2-year cardiac

mortality and easily to

calculate in clinical

application.

The ROC curves for cardiac death and MACE

The discriminatory power for

MACE was similar for all

models in our study, AUC

ranging from 0.592 to 0.609.

Limitations

First, we didn’t distinguish the different generation drug-eluting stent,

and bare metal stent were used very little in our population.

Second, this study was a single-center experience, with different

highly experienced operators, and was concluded in a homogenous

Chinese population.

Third,, the SS and all score models were initially made as an

angiographic predictor of clinical outcomes in patients with

multivessel coronary artery disease. Our study enrolled 76% patients

with single-vessel, expanded the application area of CSS.

Conclusions

All these combined risk models were designed to choose of

the best revascularization strategy, predict better prognostic

information, and can be individualized treatment rather than

divide patients into different risk categories.

Based on the ‘law of parsimony’ or ‘the Ockham razor’,

CSS was a simple and convenient prognostic tool for

clinical application, and easily to calculate, and may help to

decide the best management strategies to improve the

prognosis of ACS patients after PCI.