Embed Size (px)
Citation preview
���� ���������� �� ���� �� ������ ����������� ����� �������� � ������� ����
������ ���� ����� ������
������ ������� ����� ������� ��� �������� ����� ��������� ����� ������� ����� ������ ������ ����������� ���� ������ � !��� ������
Objective—It is not clear whether the severity ofcoronary artery disease as assessed on angiography hasan impact on the postoperative outcome after coronaryartery bypass surgery (CABG).Design—The angiographic status of 15 coronaryarteries/segments of 2233 patients who underwentisolated on-pump CABG was graded according to thefollowing criteria: 1 = no stenosis; 2 = stenosis�50%;3 = stenosis of 50–69%; 4 = stenosis of 70–89%;5 = stenosis of 90–99%; 6 = vessel occlusion; and7 = vessel is not visualized.Results—Thirty-seven patients (1.7%) died during thein-hospital stay and 108 (4.8%) developed postopera-tively low cardiac output syndrome. Multivariateanalysis showed that along with other risk factors theoverall coronary angiographic score was predictive ofpostoperative death (p = 0.03; OR: 1.027, 95% CI:1.003–1.052) and of low cardiac output syndrome(p = 0.04; OR: 1.172, 95% CI: 1.010–1.218). The statusof the proximal segment of the left circumflex coronaryartery, the diagonal arteries and the left obtusemarginal arteries was most closely associated withadverse postoperative outcome.Conclusion—The angiographic status of coronaryarteries has an impact on the immediate outcome afterCABG.
Key Words: angiography, coronary artery bypasssurgery, coronary artery disease, postoperative out-come, score
Angiography has a pivotal position in the managementof patients with coronary artery disease, as it is the mostreliable tool to ascertain the anatomical severity of sucha disease. When coronary artery bypass surgery(CABG) is indicated, coronary angiographic findingsform the basis for planning the strategy of surgicalrevascularization.
It is likely that the extent of coronary artery disease asassessed on coronary angiograms has the potential ofpredicting the postoperative outcome after CABG.However, scoring methods for prediction of outcomeafter CABG are mainly based on preoperative clinicalrisk factors which do not take into account the coronaryangiographic findings, but rather rely on comorbiditiesand on other indirect measures of the severity ofatherosclerotic disease. This could be one reason whythese models have been found to be variably inaccuratein predicting postoperative mortality after CABG (1).Furthermore, the absence of a method of quantificationof coronary artery disease prevents any estimation ofepidemiologic differences in the distribution andseverity of such a disease, and the evaluation of thepolicy and adequacy of coronary revascularization.
In the present study, we evaluated the value of acoronary angiographic scoring method to predict theimmediate postoperative outcome after CABG and toidentify those coronary arteries and their segmentswhose extent of atherosclerosis is associated with pooroutcome.
�������� ��� �������
From January 1997 to December 2001, 2630 patients underwent isolatedon-pump CABG at our institution and 45 patients (1.7%) died during thein-hospital stay.
Data on pre-, intra- and postoperative variables as well as those on theangiographic status of the coronary artery were available in 2233 patients,and these patients form the basis of the present study. Data on pre- andpostoperative outcome were collected retrospectively, whereas angio-graphic findings of the status of the coronary arteries as well as all theintraoperative variables were collected prospectively.
Left heart side catheterization was performed by Judkins’ technique.To obtain optimal assessment of each specific coronary artery/segment,the left coronary artery was examined in six projections, and the rightcoronary artery at least in two projections. The angiographic status of thecoronary arteries/segments was graded according to the followingcriteria: 1 = no stenosis; 2 = stenosis�50%; 3 = stenosis of 50–69%;4 = stenosis of 70–89%; 5 = stenosis of 90–99%; 6 = vessel occlusion;and 7 = vessel is not visualized. These scores were assigned to the
2003 Taylor & Francis.ISSN 1401–7431DOI 10.1080/14017430310014993 Scand Cardiovasc J 37
Received May 23, 2003; accepted June 18, 2003.1Divisions of Cardiothoracic and Vascular Surgery, Department of Surgery ,2Division of Cardiology, Department of Medicine, University of Oulu and Oulu UniversityHospital, Oulu, Finland.Correspondence to Fausto Biancari, MD, PhD, Division of Cardiothoracic and Vascular Surgery, Department of Surgery, Oulu University Hospital, PO Box 21, FI-90029Oulu, Finland. Tel:�358 8 34 6666. Fax:�358 8 315 2577. E-mail: [email protected] Cardiovasc J 37; 275–282, 2003
Scan
d C
ardi
ovas
c J
Dow
nloa
ded
from
info
rmah
ealth
care
.com
by
Uni
vers
ity o
f A
uckl
and
on 1
0/29
/14
For
pers
onal
use
onl
y.
following vessels/segments: on the left side, the left main coronary artery,the anterior descending artery divided into proximal, middle and distalsegments, the first and second diagonal arteries, the ramus medianus, thecircumflex artery divided into proximal, middle and distal segments, andthe first and second obtuse marginal branches; and on the right side, theright coronary artery divided into proximal, middle and distal segments.The overall coronary angiographic score was calculated by summing thescore of each of these arteries/segments. The minimum overall possiblescore is 15 and the maximum is 105.
Statistical analysis was performed using the SPSS software (SPSS v.10.0.5, SPSS Inc., Chicago, IL, USA). Continuous variables are reportedas the mean� standard deviation. The�2-test and Fisher’s exact test withor without the Monte-Carlo method were used for univariate analysis ofcategorical data. The Mann–Whitney test and the Kruskal–Wallis testwere used to assess the distribution of continuous variables in differentsubgroups. The Spearman test was used to evaluate the correlationbetween variables. The area under the receiver-operating characteristic(ROC) curve was calculated to depict the relationship between thepostoperative endpoints and coronary angiographic score, and to identifythe related cut-off points. These cut-off points were determined bychoosing the value with the highest sensitivity and specificity. Logisticregression with the help of backward selection was used for multivariateanalysis. Unless otherwise indicated, multivariate analysis included onlythe preoperative variables shown to be significantly associated with theoutcome at univariate analysis. Odds ratios for overall coronaryangiographic score, cardiac ejection fraction, and patient’s age arereported for an increase of 10 units. A value ofp � 0.05 was consideredstatistically significant.
��!"��
There were 564 women (25.3%) and 1669 men (74.7%)with a mean age of 63.5� 9.3 years. Five hundred andtwenty-four patients (23.5%) had diabetes, 1001 pa-tients (44.8%) had a history of myocardial infarction,108 (4.8%) of stroke and 162 (7.3%) of lower limbischemia. Elective CABG was performed in 1567patients (70.2%), emergency operation in 532 (23.8%)and urgent operation in 134 (6.0%). Redo CABG wasperformed in 53 patients (2.4%), one of them havingbeen a second re-operation. In 1694 cases (75.9%),patients were preoperatively in NYHA classes III–IV.The preoperative cardiac ejection fraction was known in1290 patients and its mean value was 65.9� 15.5. Themean overall coronary angiographic score was39.0� 12.1, and details on its distribution as well asthe score of each coronary artery/segment distributionamong different preoperative risk factors and post-operative outcome endpoints are reported in Table I.Data on coronary artery dominance were available in2194 cases. A right dominance was present in 1427patients (65.0%), left dominance in 369 (16.8%) andthere was a balance in 398 cases (18.1%). The meannumber of distal anastomoses performed was 3.8� 1.0.
During the postoperative period, 108 patients (4.8%)developed low cardiac output syndrome, which requiredprolonged vasoactive treatment in 84 cases (3.8%),intra-aortic balloon pump in 21 cases (0.9%), redocoronary artery bypass grafting in two cases (0.1%) andmitral valve replacement in one case (0.05%). Thirty-seven patients (1.7%) died during the in-hospital stay.
Cardiac complication was the primary cause of death in25 patients, neurological complication in 10 patients,bleeding in one case and multiorgan failure in anothercase. In almost all cases, the clinical course of thesepatients was complicated by cardiac and/or multiorganfailure.
������ �� ������� ���� �� �� ��������������Univariate analysis showed that the preoperativeNYHA class status (p � 0.0001), history of myocardialinfarction (p = 0.005), diabetes (p = 0.04), lower limbischemia (p = 0.04), the emergency and urgent nature ofthe procedure (p � 0.0001), patient’s age (p � 0.0001),cardiac ejection fraction (p � 0.0001), and the overallcoronary angiographic score (p = 0.016) were predictiveof postoperative death.
Among the operative variables, blood losses (p =0.04) and the length of perfusion (p � 0.0001), but notthe aortic cross-clamping time (p = 0.07), were pre-dictors of postoperative death.
Multivariate analysis including only the preoperativevariables showed that cardiac ejection fraction (p �0.0001; OR: 0.505, 95% CI: 0.385–0.671) and historyof lower limb ischemia (p = 0.008; OR: 4.243, 95% CI:1.458–12.352) were predictors of postoperative death.However, the inclusion of cardiac ejection fractionrestricted the analysis only to 1290 patients. When theanalysis was extended to all patients by excludingpreoperative cardiac ejection fraction from the model,multivariate analysis showed that patient’s age (p =0.01; OR: 1.174, 95% CI: 1.130–2.665), the preopera-tive NYHA class status (p = 0.001) and the coronaryangiographic score (p = 0.03; OR: 1.027, 95% CI:1.003–1.052) were predictors of postoperative death.
The best cut-off value for coronary angiographicscore in predicting postoperative death was 39.5(sensitivity: 54.5%, specificity: 61.6%; area under thecurve: 0.615, SE = 0.049,p = 0.016). Postoperativedeath occurred in 15 patients (1.1%) having anangiographic score of�40 and in 22 patients (2.5%)having a score�40 (p = 0.004; OR: 2.354, 95% CI:1.214–4.562).
������ �� ������� ���� �� �� �� ���������� ���������� ���������� ������Univariate analysis showed that female sex (p = 0.01),the preoperative NYHA class status (p � 0.0001),history of myocardial infarction (p � 0.0001), theemergency and urgency condition of the procedure(p � 0.0001), redo operation (p = 0.004), age (p �0.0001), preoperative cardiac ejection fraction(p � 0.0001), and the overall coronary angiographicscore (p = 0.001) were predictive of the development ofpostoperative low-output syndrome.
Scand Cardiovasc J 37
276 F. Biancari et al.
Scan
d C
ardi
ovas
c J
Dow
nloa
ded
from
info
rmah
ealth
care
.com
by
Uni
vers
ity o
f A
uckl
and
on 1
0/29
/14
For
pers
onal
use
onl
y.
Tab
leI.
Cor
onar
yan
giog
raph
icsc
ores
acco
rdin
gto
diff
eren
tri
skfa
ctor
san
dpo
stop
erat
ive
outc
ome
endp
oint
s
No.
Left
mai
ntr
unk
LAD
aLA
Db
LAD
cD
iag.
1D
iag.
2IM
CX
aC
Xb
CX
cLO
M1
LOM
2R
CA
aR
CA
bR
CA
c
Ove
rall
angi
ogra
phic
scor
e
Pre
oper
ativ
eri
skfa
ctor
sM
yoca
rdia
lin
farc
tion
No
1232
(55.
2)1.
8**
3.6*
*2.
81.
82.
8**
2.0
2.0*
*2.
7**
2.6
1.8
2.6
2.0
3.2*
3.1*
**2.
7**
37.7
***
Yes
1001
(44.
8)1.
63.
73.
02.
13.
12.
22.
22.
92.
61.
92.
72.
23.
43.
63.
140
.5D
iabe
tes
No
1709
(76.
5)1.
8*3.
72.
92.
02.
92.
12.
0*2.
7**
2.5*
*1.
82.
62.
13.
2**
3.3
2.9
38.6
**Y
es52
4(2
3.5)
1.6
3.6
2.9
2.0
3.0
2.1
2.2
3.0
2.8
2.0
2.8
2.2
3.5
3.5
2.9
40.2
Str
oke
No
2125
(95.
2)1.
73.
62.
92.
02.
9**
2.1
2.1
2.8*
*2.
61.
92.
62.
13.
3*3.
32.
9*38
.8**
Yes
108
(4.8
)1.
83.
53.
22.
22.
92.
52.
23.
22.
62.
12.
82.
43.
63.
73.
341
.9Lo
wer
limb
isch
emia
No
2071
(92.
7)1.
7**
3.7
2.9
2.0
2.9
2.1
2.1
2.8*
2.6
1.9
2.7
2.1
3.2*
**3.
3***
2.9
38.9
Yes
162
(7.3
)1.
93.
42.
71.
82.
91.
91.
93.
12.
62.
02.
62.
04.
04.
03.
040
.0N
YH
Acl
ass
I28
(1.3
)1.
3**
3.5*
**2.
31.
62.
72.
1*1.
92.
6**
2.4
1.5*
2.4
1.7
2.8*
2.8
2.5
34.2
***
II51
1(2
2.9)
1.6
3.4
2.9
1.9
2.9
2.0
2.0
2.6
2.5
1.8
2.6
2.1
3.1
3.2
2.8
37.4
III11
18(5
0.1)
1.7
3.7
2.9
2.0
2.9
2.0
2.0
2.8
2.6
1.9
2.6
2.1
3.4
3.4
2.9
38.9
IV57
6(2
5.8)
1.8
3.8
3.0
2.1
3.1
2.4
2.3
3.0
2.6
1.8
2.8
2.1
3.3
3.4
3.0
40.6
Typ
eof
oper
atio
nE
lect
ive
1567
(70.
2)1.
6***
3.6*
*3.
02.
02.
92.
12.
12.
72.
61.
92.
62.
13.
33.
32.
938
.7U
rgen
t53
2(2
3.8)
2.0
3.8
2.8
1.9
3.0
2.1
2.0
2.9
2.6
2.0
2.6
2.0
3.3
3.3
2.9
39.4
Em
erge
nt13
4(6
.0)
1.8
3.9
3.0
2.0
2.9
2.5
2.4
3.0
2.5
1.8
2.8
2.1
3.4
3.2
2.8
40.1
Firs
tC
AB
GN
o21
80(9
7.6)
1.7
3.6*
**2.
92.
02.
9**
2.1*
2.1
2.8*
**2.
61.
92.
6**
2.1
3.3*
**3.
32.
938
.8**
*R
edo
CA
BG
Yes
53(2
.4)
1.9
4.7
3.1
2.1
3.9
2.9
2.5
3.8
2.7
2.3
3.7
2.6
4.4
3.8
2.9
47.5
Cor
onar
ydo
min
ance
Bal
ance
d39
8(1
8.1)
1.7
3.3*
**2.
81.
82.
92.
02.
02.
62.
6*1.
92.
62.
13.
5***
3.7*
*3.
2***
38.9
Rig
ht14
27(6
5.0)
1.8
3.7
3.0
2.0
2.9
2.1
2.1
2.9
2.7
1.9
2.6
2.1
3.1
3.3
2.9
39.0
Left
369
(16.
8)1.
73.
92.
92.
13.
12.
42.
22.
82.
31.
92.
72.
13.
53.
22.
539
.2O
utco
me
Pos
tope
rativ
ede
ath
No
2196
(98.
3)1.
73.
62.
92.
02.
9*2.
1*2.
12.
8**
2.6
1.9
2.6*
*2.
13.
33.
32.
938
.9*
Yes
37(1
.7)
1.5
3.9
2.6
2.3
3.9
3.0
2.5
3.6
3.0
2.2
3.8
2.6
3.7
3.4
2.9
44.8
Low
-out
put
synd
rom
eN
o21
25(9
5.2)
1.7
3.6*
*2.
92.
02.
9**
2.1*
*2.
0*2.
8**
2.6
1.9
2.6
2.1
3.3*
3.3
2.9
38.7
**Y
es10
8(4
.8)
1.7
4.0
2.9
2.1
3.5
2.9
2.6
3.4
2.7
1.9
3.1
2.4
3.7
3.3
2.7
42.9
LAD
=le
ftan
terio
rde
scen
ding
coro
nary
arte
ry;
a=
prox
imal
segm
ent;
b=
mid
dle
segm
ent;
c=
dist
alse
gmen
t;D
iag.
=di
agon
alar
tery
;C
X=
left
circ
umfle
xco
rona
ryar
tery
;LO
M=
left
obtu
sem
argi
nal
arte
ry;
RC
A=
right
coro
nary
arte
ry.
*p�
0.05
;**
p�
0.01
;**
*p�
0.00
01.
Scand Cardiovasc J 37
Coronary angiographic score 277
Scan
d C
ardi
ovas
c J
Dow
nloa
ded
from
info
rmah
ealth
care
.com
by
Uni
vers
ity o
f A
uckl
and
on 1
0/29
/14
For
pers
onal
use
onl
y.
Among the intraoperative variables, blood loss(p = 0.02), the length of perfusion (p � 0.0001) andthe aortic cross-clamping time (p = 0.01) were predic-tors of postoperative low-output syndrome.
Multivariate analysis of the preoperative variablesshowed that cardiac ejection fraction (p � 0.0001; OR:0.693, 95% CI: 0.580–0.0.817) and the urgent/emergentnature of the operation (p = 0.001) were predictors ofpostoperative low-output syndrome. The inclusion ofcardiac ejection fraction restricted the analysis only to1290 patients. When the analysis was extended to allpatients by excluding preoperative cardiac ejectionfraction from the model, multivariate analysis showedthat patient’s age (p � 0.0001; OR: 1.582, 95% CI:1.243–2.004), the urgent/emergent nature of the opera-tion (p � 0.0001), redo operation (p = 0.003; OR: 3.763,95% CI: 1.571–9.013), history of myocardial infarction(p = 0.01; OR: 1.616, 95% CI: 1.063–2.456), and thecoronary angiographic score (p = 0.04; OR: 1.172, 95%CI: 1.010–1.218) were predictors of postoperative low-output syndrome.
The best cut-off value for coronary angiographicscore in predicting postoperative low-output syndromewas 39.5 (sensitivity: 53.7%, specificity: 62.0%; areaunder the curve: 0.593, SE = 0.029,p = 0.001). Post-operative low-output syndrome occurred in 50 patients(3.6%) having an angiographic score of�40 and in 58patients (6.7%) having a score�40 (p = 0.001; OR:1.894, 95% CI: 1.285–2.793).
������ ����������� ��� ��� ������������� �������Coronary angiographic score was significantly corre-lated with the preoperative value of cardiac ejectionfraction (p = 0.122, p � 0.0001). The mean angio-graphic score was 38.7� 11.4 among patients with anejection fraction�50% and 42.7� 13.2 among thosewith an ejection fraction�50% (p � 0.0001).
����������� ������ �� ��� ������ ��������� ��� ���������� ������The results of univariate analysis for prediction ofpostoperative outcome according to the angiographicstatus of each coronary artery/segment are reported inTable I. Multivariate analysis including only thecoronary artery/segments which were found to besignificantly associated with postoperative outcome atunivariate analysis, showed that the angiographic statusof the first diagonal artery (p = 0.01; OR: 1.256, 95%CI: 1.057–1.492) and of the first left obtuse marginalartery (p = 0.002; OR: 1.278, 95% CI: 1.090–1.498)were predictive of postoperative death, whereas theangiographic status of the second diagonal artery(p = 0.001; OR: 1.140, 95% CI: 1.052–1.232) and ofthe proximal segment of the left circumflex artery
(p = 0.002; OR: 1.201, 95% CI: 1.072–1.345) werepredictive of postoperative low cardiac output syn-drome.
����������� ������ �� �� ������� ����� �� ����� ������ ������� ��� ��� ����������������The degree of stenosis of the proximal segment of theleft circumflex coronary artery was significantly asso-ciated with most of the preoperative risk factors andwith the postoperative outcome endpoints (Table I).There was a correlation between the angiographic statusof this artery segment and the overall angiographicscore (p � 0.0001), in particular, there was a relevantincrease in the overall angiographic score in the case ofits occlusion or no visualization (Fig. 1). Indeed, theangiographic score of the proximal segment of the leftcircumflex coronary artery was shown to be signifi-cantly associated with the postoperative outcome (Figs2 and 3). In particular, the postoperative mortality andlow cardiac output syndrome rates, when the proximalsegment of the left coronary circumflex artery wasvisualized as open, were 1.4 and 4.3%, respectively,whereas they were 4.2 and 10.8%, respectively, when itwas occluded or not visualized (p = 0.02,p = 0.001).
When the degree of stenosis of the proximal segmentof the left circumflex artery was included in themultivariate analysis substituting the overall angio-graphic score (preoperative cardiac ejection fractionexcluded), patient’s age (p = 0.005; OR: 1.825, 95% CI:1.195–2.788), the preoperative NYHA class (p = 0.001)and the degree of stenosis involving the proximalsegment of the left circumflex artery (p = 0.03; OR:1.245, 95% CI: 1.025–1.512) were the predictors ofpostoperative death.
���� � "�����# �$��$����� ���� ���$ ������� %��� �&&���� �$���� �& ������� ������$ ��� ���'���� ��$��� �&��� ��&� ����(�' �����# �����# )�� *+***�,+
Scand Cardiovasc J 37
278 F. Biancari et al.
Scan
d C
ardi
ovas
c J
Dow
nloa
ded
from
info
rmah
ealth
care
.com
by
Uni
vers
ity o
f A
uckl
and
on 1
0/29
/14
For
pers
onal
use
onl
y.
Similarly, when the variable LCXa was included inthe multivariate analysis substituting the overall angio-graphic score, patient’s age (p � 0.0001; OR: 1.582,95% CI: 1.243–2.023), the emergent and urgent natureof the operation (p � 0.0001), redo operation (p =0.003; OR: 3.841, 95% CI: 1.613–9.144), history ofmyocardial infarction (p = 0.02; OR: 1.624, 95% CI:1.069–2.468), and the degree of stenosis involving theproximal segment of the left circumflex artery (p = 0.03;OR: 1.141, 95% CI: 1.015–1.282) were predictors ofpostoperative low-output syndrome.
������ ��� �������� ��� ����������������The postoperative mortality rate among patients with
right, balanced and left coronary dominance was 1.3,1.3 and 3.2%, respectively (p = 0.04), while the post-operative rate of low cardiac output syndrome was 3.8,6.5 and 7.3%, respectively (p = 0.003). The overallcoronary angiographic score was similar in each of thedominance patterns (p = 0.38). The angiographic scoresfor each coronary artery/segment in patients who diedor developed low cardiac output syndrome postopera-tively according to the pattern of coronary dominanceare reported in Table II.
Multivariate analysis including the angiographicscores of all coronary artery/segments and the coronarydominance pattern, showed that the angiographic statusof the first diagonal artery (p = 0.003; OR: 1.312, 95%CI: 1.096–1.572), the first left obtuse marginal artery(p = 0.004; OR: 1.268, 95% CI: 1.078–1.490), andcoronary dominance pattern (p = 0.05) were predictiveof postoperative death.
The angiographic status of the second diagonalartery (p = 0.001; OR: 1.141, 95% CI: 1.055–1.234),the proximal segment of the left circumflex artery(p = 0.002; OR: 1.200, 95% CI: 1.069–1.347), andcoronary dominance pattern (p = 0.003) were pre-dictive of postoperative low cardiac output syn-drome. These results are rather similar to those ofmultivariate analysis including only the coronaryartery/segments which were found to be significantlyassociated with postoperative outcome at univariateanalysis.
In order to better evaluate the impact of the severityof coronary artery disease affecting the major arteriesaccording to the coronary dominance pattern, theangiographic score of each segment of the rightcoronary, left anterior descending and circumflexartery as well as their branches (first and seconddiagonal for the left anterior descending artery, andfirst and second obtuse marginal for the circumflexartery) were summed to get the overall angiographicscore of three major myocardial territories. These threeregional angiographic scores as well as the angio-graphic score of the left main coronary artery and thecoronary dominance pattern were included in theregression model in order to identify the coronaryterritory having the major impact on the postoperativeoutcome. Multivariate analysis showed that the leftcircumflex coronary artery territory (p = 0.002; OR:1.070, 95% CI: 1.025–1.116) and the coronarydominance pattern (p = 0.02) were predictive of post-operative death, whereas the left anterior descendingcoronary artery territory (p = 0.02; OR: 1.034, 95% CI:1.006–1.063), the left circumflex coronary arteryterritory (p = 0.02; OR: 1.036, 95% CI: 1.006–1.066)and the coronary dominance pattern (p = 0.002) werepredictive of postoperative low cardiac output syn-drome.
���� !� ������������� ��������# ����� ��� �$ �� �&&���� �$���� �& ������� ������$ ��� ���'���� ��$��� �& ��� ��&�����(�' �����# �����# )� - *+*�,+ !�� ��.�� �& ������� &����� �$��� �& ������� �� ������� � ��$+ �+
���� "� ������������� ��% �� �� ������ �# ���� ����� ��� /�$ �� �&&���� �$���� �& ������� ������$ ��� ���'���� ��$����& ��� ��&� ����(�' �����# �����# )� - *+*�,+ !�� ��.�� �&������� &�� ��� �$��� �& ������� �� ������� � ��$+ �+
Scand Cardiovasc J 37
Coronary angiographic score 279
Scan
d C
ardi
ovas
c J
Dow
nloa
ded
from
info
rmah
ealth
care
.com
by
Uni
vers
ity o
f A
uckl
and
on 1
0/29
/14
For
pers
onal
use
onl
y.
������ ����������� ��� ��� ��#���� �������� ��������$�����The mean number of distal anastomoses in patients whodied postoperatively was 3.7� 1.0, whereas it was3.8� 0.8 among survivors (p = 0.38). The mean num-ber of distal anastomoses was the same among patientswho suffered postoperatively of low cardiac outputsyndrome and in those who did not (3.8,p = 0.8).Figures 4 and 5 show the mean number of distalanastomoses according to different tertiles of the
Tab
leII.
Mea
nco
rona
ryan
giog
raph
icsc
ores
inpa
tien
tsw
hodi
edor
deve
lope
dlo
wca
rdia
cou
tput
synd
rom
epo
stop
erat
ivel
yac
cord
ing
toco
rona
rydo
min
ance
No.
(%)
Left
mai
ntr
unk
LAD
aLA
Db
LAD
cD
iag.
1D
iag.
2IM
CX
aC
Xb
CX
cLO
M1
LOM
2R
CA
aR
CA
bR
CA
c
Ove
rall
angi
ogra
phic
scor
e
Pos
tope
rati
vede
ath
Rig
htdo
min
ance
Aliv
e14
08(9
8.7)
1.8*
3.7
3.0
2.0
2.9
2.1
2.1*
2.8
2.6
1.9
2.6
2.1*
3.1
3.3
2.9
38.9
Dea
th19
(1.3
)1.
23.
82.
82.
13.
82.
72.
93.
43.
22.
03.
63.
13.
83.
62.
944
.9B
alan
ced
dom
inan
ceA
live
393
(98.
7)1.
73.
32.
81.
82.
92.
02.
02.
6*2.
61.
92.
62.
13.
53.
73.
138
.8D
eath
5(1
.3)
2.0
3.0
1.6
1.6
3.8
2.2
1.8
4.4
3.6
2.2
4.4
3.4
4.2
3.6
3.8
45.6
Left
dom
inan
ceA
live
357
(96.
7)1.
63.
92.
92.
13.
02.
3*2.
22.
72.
31.
92.
7*2.
13.
53.
22.
539
.1D
eath
12(3
.3)
1.8
4.4
2.8
2.8
4.2
4.0
2.2
3.4
2.2
2.7
3.7
2.5
3.1
3.2
2.7
44.7
Pos
tope
rati
velo
wca
rdia
cou
tput
synd
rom
eR
ight
dom
inan
ceN
o13
75(9
3.5)
1.8
3.6*
*3.
02.
02.
92.
12.
12.
92.
71.
92.
62.
13.
13.
32.
938
.9Y
es52
(6.5
)1.
54.
22.
71.
93.
52.
52.
33.
32.
71.
73.
02.
53.
43.
32.
540
.9B
alan
ced
dom
inan
ceN
o37
2(9
6.4)
1.7
3.3
2.8
1.8
2.8
2.0
2.0
2.6
2.6
1.9
2.6
2.0
3.5
3.7
3.1
38.4
*Y
es26
(3.6
)1.
83.
62.
82.
03.
52.
82.
63.
03.
12.
63.
22.
93.
93.
73.
545
.2Le
ftdo
min
ance
No
342
(92.
7)1.
63.
92.
92.
13.
02.
3**
2.1*
2.7*
*2.
31.
92.
72.
13.
53.
22.
538
.8*
Yes
27(7
.3)
2.0
4.1
3.1
2.4
3.7
3.9
3.4
3.6
2.6
1.9
3.2
2.0
3.7
3.1
2.7
45.1
LAD
=le
ftan
terio
rde
scen
ding
coro
nary
arte
ry;
a=
prox
imal
segm
ent;
b=
mid
dle
segm
ent;
c=
dist
alse
gmen
t;D
iag.
=di
agon
alar
tery
;C
X=
left
circ
umfle
xco
rona
ryar
tery
;LO
M=
left
obtu
sem
argi
nal
arte
ry;
RC
A=
right
coro
nary
arte
ry.
*p�
0.05
;**
p�
0.01
.
���� %���� ��.�� �& ����� ���������� ��� �$ �� �&&������������ �& �����# �$��$����� ���� � ������� %�� ������� � � ����� %�� �� ���$ ��� �/�������� ���# )� ��� ��������� *+��,+
���� &���� ��.�� �& ����� ���������� ��� �$ �� �&&������������ �& �����# �$��$����� ���� � ������� %�� �� � ������ %�� � �� ���� ��������������# ��% �� �� �������# ���� )� ��� ������� �� *+�0,+
Scand Cardiovasc J 37
280 F. Biancari et al.
Scan
d C
ardi
ovas
c J
Dow
nloa
ded
from
info
rmah
ealth
care
.com
by
Uni
vers
ity o
f A
uckl
and
on 1
0/29
/14
For
pers
onal
use
onl
y.
coronary angiographic score in patients who died or hadpostoperatively low cardiac output syndrome and inthose who did not. No statistically significant differ-ences were observed in each tertile of the overallcoronary angiographic score between patients who died(p � 0.11) or developed postoperatively low cardiacoutput syndrome (p � 0.28) and those who were notaffected by such adverse events.
���#!�����
The search for risk factors associated with unfavorableoutcome is of main importance not only for riskstratifying purposes, but also for better planning oftreatment strategies focusing on those patients at higherrisk of adverse outcome. Current clinical risk scoreshave been developed with such aims and in cardiacsurgery they serve also for control of quality of surgicalcenters. Indeed, unlike other fields of surgery, thepreoperative risk stratification in this patient populationis seldom associated with cancellation of operation inthose patients with the highest operative risk, and oftenthe preoperative optimization of the general status of thepatients can not be afforded. Thus, optimization of theanesthesiologic management and of the operativestrategy is of key importance in cardiac surgery. Thisis particularly true in patients undergoing CABG as theoptimal surgical strategy for coronary artery disease isstill a matter of debate (2, 3). It is objectively difficult toestablish what is a complete or effective revasculariza-tion as it is not clear what is the impact of the severity ofcoronary artery disease on the immediate and long-termoutcome. A few large studies addressed this topic andconcluded that the angiographic status of the coronariesis of major prognostic importance (4–9), irrespective ofthe type of treatment (6, 8). Because of this, one of thesecoronary angiographic scores, the Duke score, has beenused to define at which extent of coronary artery diseaseCABG provides better results than percutaneous trans-luminal angioplasty (6, 8).
Herein, we employed a rather descriptive scoringmethod that leaves unweighted any artery/segment andgrades the degree of stenosis in the same fashion for allthe vessels. This permitted the evaluation of the impactof the overall angiographic score and the identificationof those vessels and their segments being associatedwith poor prognosis. In this sense, the present scoringmethod is less restrictive than others previously usedwhich are somewhat weighted toward the severity of theleft main coronary artery (4, 7) or the left anteriordescending coronary artery (6). Furthermore, angio-graphic scoring methods taking into account only themain coronary arteries and those falling into the one-,two- and three-vessel disease classification are notdescriptive enough as they do not consider important
branches such as the diagonal arteries, the ramusmedianus and the left obtuse marginal arteries. Indeed,in this study the latter vessels were found to be of greatprognostic importance at much larger extent than theleft main and the left anterior descending coronaryartery.
This observation along with the findings on theimpact of the angiographic status of the proximalsegment of the left circumflex coronary artery on theimmediate postoperative outcome brought some indir-ect evidence on the importance of the territory ofmyocardium supplied by these vessels. Although we donot have data regarding the presence and the degree ofmitral regurgitation in this patient series, the prognosticvalue of the diagonal, left obtuse marginal and proximalsegment of the left circumflex artery suggests thatsevere coronary artery disease affecting this vesselmight likely have affected the function of the mitralapparatus, i.e. the left ventricle, papillary muscles andannulus. Ischemic mitral regurgitation is particularlyassociated with posterior infarction rather than anteriorinfarction (10–14). Sharma et al. (14) observed that80% of patients with acute severe ischemic mitralregurgitation had a stenosis�70% of the right and leftcircumflex coronary arteries with or without stenosis ofthe left anterior descending coronary artery. Ko et al.(15) showed that in patients with a stenosis�60% of theleft main coronary artery undergoing coronary angio-graphy, severe left circumflex coronary artery diseaseand NYHA class III–IV were independent predictors ofdeath after angiography. However, also these authorsdid not provide data on mitral valve function which maysupport the hypothesis that significant changes in themitral valve competence, likely associated with severestenosis of the left circumflex coronay artery, may haveadversely affected the mitral apparatus.
Elhendy et al. (16) showed that myocardial viabilityand functional recovery after CABG are not related tothe severity of coronary stenosis or the grade ofcollateral circulation. Their results suggest that surgicalrevascularization distally to an occluded coronary arteryis associated with good function outcome. However,other authors observed that non-viable segments weremore likely to be located distally to an occluded vesseland less collateralized than viable segments (17, 18).The present finding of significantly increased risk ofpostoperative death and low cardiac output syndromeafter CABG in patients with occluded proximal segmentof the left circumflex coronary artery is more in linewith the latter point of view. Although we do not haveany data to quantify the amount of collateral vessels, theassociation of significantly increased overall coronaryangiographic score in patients with occluded proximalsegment of the left circumflex coronary artery indirectlysuggests a poorer collateral circulation in this coronary
Scand Cardiovasc J 37
Coronary angiographic score 281
Scan
d C
ardi
ovas
c J
Dow
nloa
ded
from
info
rmah
ealth
care
.com
by
Uni
vers
ity o
f A
uckl
and
on 1
0/29
/14
For
pers
onal
use
onl
y.
artery disease pattern. It is likely that our ratheraggressive revascularization strategy as suggested bythe relatively high mean number of distal anastomoses,despite attempting to complete myocardial revascular-ization, was not associated with adequate recovery ofmyocardial function in critical territories as supplied bythe left circumflex, left obtuse marginal and diagonalarteries.
In conclusion, the degree of stenosis involving themajor coronary arteries as assessed on angiographyseems to have an important impact on the immediatepostoperative outcome after CABG. In particular, thestatus of the proximal segment of the left circumflexcoronary artery, the diagonal arteries and the left obtusemarginal arteries is significantly associated with most ofthe preoperative risk factors, and with postoperativemortality and the development of low cardiac outputsyndrome. Further studies are required to evaluatewhether and at what extent complete surgical revascu-larization of these territories is associated with func-tional myocardial recovery and improved immediatepostoperative outcome.
�#$��%"��&�����
Dr F. Biancari was supported by a grant of the Einar and KarinStroems Foundation which is kindly acknowledged.
�'� ��#��
1. Pinna-Pintor P, Bobbio M, Colangelo S, et al. Inaccuracy of four coronarysurgery risk-adjusted models to predict mortality in individual patients. Eur JCardiothorac Surg 2002; 21: 199–204.
2. van der Brand MJBM, Rensing BJWM, et al. The effect of completeness ofrevascularization on event-free survival at one year in the ARTS trial. J Am CollCardiol 2002; 39: 559–564.
3. Vander der Salm TJ, Kip KE, Jones RH, et al. What constitutes optimal surgicalrevascularization? Answers from the Bypass Angioplasty RevascularizationInvestigation (BARI). J Am Coll Cardiol 2002; 39: 565–572.
4. Gersh BJ, Kronmal RA, Frye RL, et al. Coronary arteriography and coronaryartery bypass surgery: Morbidity and mortality in patients ages 65 years orolder. A report from the Coronary Artery Surgery Study. Circulation 1983; 67:483–491.
5. Graham MM, Chambers RJ, Davies RF. Angiographic quantification of diffusecoronary artery disease: Reliability and prognostic value for bypass operations.J Thorac Cardiovasc Surg 1999; 118: 618–627.
6. Jones RH, Kesler K, Phillps HR 3rd, et al. Long-term survival benefits ofcoronary artery bypass grafting and percutaneous transluminal angioplasty inpatients with coronary artery disease. J Thorac Cardiovasc Surg 1996; 111:1013–1025.
7. Kennedy JW, Kaiser GC, Fisher LD, et al. Clinical and angiographic predictorsof operative mortality from the collaborative study in coronary artery surgery(CASS). Circulation 1981; 63: 793–802.
8. Kurbaan AS, Rickards AF, Ilsley CDJ, et al. Relation between coronary arterydisease, baseline clinical variables, revascularization mode, and mortality.CABRI Participants. Coronary Angioplasty vs. Bypass RevascularisationInvestigation. Am J Cardiol 2000; 86: 938–942.
9. Ringqvist I, Fisher LD, Mock M, et al. Prognostic value of angiographic indicesof coronary artery disease from the Coronary Artery Surgery Study (CASS). JClin Invest 1983; 71: 1854–1866.
10. Heikkila J. Mitral incompetence as a complication of acute myocardialinfarction. Acta Medica Scand Suppl 1967; 475: 1–149.
11. Gorman JH 3rd, Gorman RC, Plappert T, et al. Infarct size and locationdetermine development of mitral regurgitation in the sheep model. J ThoracCardiovasc Surg 1998; 115: 615–622.
12. Gorman JH 3rd, Gorman RC, Jackson BM, et al. Distortions of the mitral valvein acute ischemic mitral regurgitation. Ann Thorac Surg 1997; 64: 1026–1031.
13. Edmunds LH Jr. Ischemic mitral regurgitation. In: Edmunds LH Jr, editor.Cardiac surgery in the adult. New York: McGraw-Hill 1997: 657–676.
14. Sharma SK, Seckler J, Israel DH, et al. Clinical, angiographic and anatomicfindings in acute severe ischemic mitral valve regurgitation. Am J Cardiol 1992;70: 277–280.
15. Ko JK, Nishimura RA, Holmes DR Jr, et al. Predictors of early mortality inpatients with angiographically documented left main coronary artery disease.Catheter Cardiovasc Diagn 1991; 24: 84–87.
16. Elhendy A, Cornel JH, Roelandt JRTC, et al. Impact of severity of coronaryartery stenosis and the collateral circulation on the functional outcome ofdyssynergic myocardium after revascularization in patients with healedmyocardial infarction and chronic left ventricular dysfunction. Am J Cardiol1997; 79: 883–888.
17. Di Carli M, Sherman T, Khanna S, et al. Myocardial viability in asynergicregions subtended by occluded coronary arteries: Relation to the status of thecollateral flow in patients with chronic coronary artery disease. J Am CollCardiol 1994; 23: 860–868.
18. Marzullo P, Parodi O, Sambuceti G, et al. Residual coronary reserve identifiessegmental viability in patients with wall motion abnormalities. J Am CollCardiol 1995; 26: 342–350.
Scand Cardiovasc J 37
282 F. Biancari et al.
Scan
d C
ardi
ovas
c J
Dow
nloa
ded
from
info
rmah
ealth
care
.com
by
Uni
vers
ity o
f A
uckl
and
on 1
0/29
/14
For
pers
onal
use
onl
y.
