Keywords
chronic total occlusion - CTO - complete revascularization - CABG - surgical revascularization
Introduction
Chronic total occlusions (CTOs) are a common finding in patients with coronary artery
disease (CAD) and occur with a prevalence of 11 to 35% in patients with CAD.[1]
[2]
[3]
[4] Coronary collaterals are seen in most CTO patients in variable occurrence and extent.
Cohen and Rentrop classification is used to grade these collaterals into four different
categories using the visibility of the collaterals as well as the occluded vessel
for scoring ([Fig. 1]).[5]
Table 1
Patient characteristics based on Cohen and Rentrop grade of collateral filling
|
Rentrop 0 (n = 23)
|
Rentrop 1 (n = 85)
|
Rentrop 2 (n = 141)
|
Rentrop 3 (n = 155)
|
p-Value
|
|
Age (y) (mean ± SD)
|
72.26 ± 6.283
|
66.16 ± 9.209
|
65.79 ± 9.569
|
67.50 ± 8.853
|
0.011
|
|
Male (%)
|
22 (95.7)
|
78 (91.8)
|
114 (81.4)
|
140 (90.3)
|
0.034
|
|
Hypertension (%)
|
16 (69.6)
|
73 (85.9)
|
112 (80)
|
126 (81.3)
|
0.321
|
|
Hyperlipidemia (%)
|
14 (60.9)
|
54 (63.5)
|
93 (66.4)
|
102 (65.8)
|
0.95
|
|
Smoker (%)
|
6 (26.1)
|
32 (37.6)
|
59 (42.1)
|
52 (33.5)
|
0.333
|
|
Diabetes mellitus (%)
|
10 (43.5)
|
26 (30.6)
|
50 (35.5)
|
40 (25.8)
|
0.177
|
|
COPD (%)
|
2 (8.6)
|
6 (7.1)
|
19 (13.5)
|
18 (11.7)
|
0.497
|
|
Peripheral artery disease (%)
|
1 (4.3)
|
15 (17.6)
|
34 (24.3)
|
24 (15.5)
|
0.071
|
|
Cerebral artery disease (%)
|
6 (26.1)
|
21 (24.7)
|
25 (17.9)
|
26 (16.8)
|
0.372
|
|
Prior apoplex (%)
|
1 (4.3)
|
7 (8.2)
|
14 (9.9)
|
20 (12.9)
|
0.492
|
|
Creatinine clearance (mL/min)
|
65.26 ± 19.845
|
91.42 ± 30.35
|
85.77 ± 32.24
|
81.8 ± 32.38
|
0.003
|
|
LVEF (%)
|
42.48 ± 10.80
|
48.74 ± 14.00
|
47.94 ± 13.51
|
47.89 ± 13.78
|
0.267
|
|
Prior PCI plus stent (%)
|
4 (17.4)
|
18 (21.2)
|
25 (17.7)
|
21 (13.5)
|
0.49
|
Abbreviations: COPD, chronic obstructive pulmonary disease; LVEF, left ventricular
ejection fraction; PCI, percutaneous coronary intervention; SD, standard deviation.
Coronary collateral circulation has an important standing in CAD as it is thought
to maintain myocardial viability and to prevent irreversible myocardial injury.[4]
[6] Apart from that, good collateralization opens up options for retrograde percutaneous
coronary intervention (PCI) strategies, whereas poor collateralization also seems
to reduce chances of successful antegrade and retrograde PCIs.[7]
Werner et al evaluated the limitation of collateral circulation showing that even
in well-collateralized patients without prior myocardial infarction (MI), revascularization
is probably beneficial.[8] Further studies described the benefits of revascularizing CTOs using PCI or coronary
artery bypass grafting (CABG), reducing all-cause mortality, if doing so.[9] Especially, aiming at complete revascularization (CR) is known to improve the patient's
outcome.[10]
[11]
[12]
[13]
CABG surgery has been described as the most successful method to fulfill CR and is
therefore specified as the optimal treatment for patients with complex CAD.[14]
[15]
The strategy for surgical revascularization usually depends on coronary dimensions
as well as complexity and severity of CAD which is analyzed in preoperative angiograms.
Considering that, the relation of preoperatively evaluated scarcely or invisibly chronic
total occluded vessels and the revascularization rate and bypass quality is not well
described in the current literature.
The purpose of this study is to evaluate if coronary visibility and grade of coronary
collateralization in preoperative angiograms reflect intraoperative findings and whether
these should influence the decision making regarding revascularization strategies.
In the present study, we focused on analyzing the visibility and collateralization
of CTO according to Cohen and Rentrop grade and evaluated successful revascularization
rate, intraoperative bypass flow, and vessel size.
Patients and Methods
Patient Population
We identified 404 patients who underwent isolated on-pump CABG surgery at our center
between 2014 and 2016 and were seen to have at least one CTO in the preoperative angiogram
([Table 1]).
Clinical and angiographic patient characteristics were retrospectively analyzed. Data
collection included the evaluation of patient electronic records regarding patient
characteristics, cardiac risk factors, detailed surgery information, preoperative
risk scores, scores evaluating CAD complexity, and clinical outcome.
Preoperative coronary angiograms were reviewed, screened for CTOs and collateralization
was classified on the basis of Cohen and Rentrop collateral recipient filling grade.
Patients who underwent prior CABG or emergency CABG due to coronary dissection during
PCI were excluded from this study. Further exclusion criteria were off-pump procedures
and any other additional surgical procedure other than CABG.
Apart from that, patients were included in which the left internal mammary artery
(LIMA) was used as a graft to revascularize the left anterior descending artery (LAD)
and venous grafts, primarily the great saphenous vein, were used as single vessels
to revascularize the left circumflex artery (LCX) and right coronary artery (RCA)
territories.
Definitions and Procedures
CTO of a coronary artery was defined as total luminal diameter stenosis, Thrombolysis
in Myocardial Infarction (TIMI) Study Group Flow Grading of zero, and known or assumed
existence of the occlusion for more than 3 months.
Coronary collateral vessels were classified by visual assessment following the grading
system published by Cohen and Rentrop ([Fig. 1]).[5]
Fig. 1 Schematic and coronary angiogram examples for grading according to Cohen and Rentrop
grade of collateral filling.
The complexity of CAD was assessed preoperatively using the anatomical SYNTAX Score[16] and SYNTAX II Score—a tool combining clinical and anatomical factors.[17]
As a measurement for surgical risk of mortality and morbidity, EuroSCORE I and Society
of Thoracic Surgery (STS) Adult Cardiac Surgery risk score were calculated.
LAD territory CTO comprises an occlusion in the LAD and major diagonal branches, whereas
an occlusion in the LCX domain and major marginal or intermedial branches was classified
as LCX territory CTO. RCA territory CTO involves occlusions in the RCA and its distal
branches—the right posterior descending artery and right posterior lateral artery.
Left ventricular (LV) function was classified using preoperative echocardiography
data and defined as normal LV function if ejection fraction (EF) was ≥ 50%. An EF
30 to 49% was defined as mildly reduced LV function and <30% as severely reduced LV
function.
Kidney disease was classified using creatinine clearance. A preoperative creatinine
clearance < 60 mL/min was defined as chronic kidney disease.
To evaluate bypass quality, mean graft flow was measured by transit time flow measurement
(TTFM) in all bypass grafts. Flow values were measured intraoperatively after weaning
from cardiopulmonary bypass and hemodynamic stabilization using the Medistim QuickFit
probes for TTFM.
Coronary diameter was assessed intraoperatively using 0.5, 1, 1.5, and 2 mm vessel
probes.
Statistical Analysis
Statistical analysis was performed using SPSS Statistics (version 24, IBM, Armonk,
New York, United States). [Figs. 1] and [2] are designed using Adobe Illustrator CC 2019 (Adobe Systems Software Ireland Limited,
Dublin, Republic of Ireland). [Figs. 3] and [4] are designed using GraphPad-Prism version 7.0 (La Jolla, California, United States).
Fig. 2 Distribution of different CTO locations. Patients predominantly had one occluded
vessel. Most frequently affecting the RCA (41.7%). CTO, chronic total occlusion; LAD,
left anterior descending artery; LCX, left circumflex artery; RCA, right coronary
artery.
Fig. 3 Percentage of the revascularized occluded vessels depending on Rentrop grading. In
occluded LAD- and LCX-dependent vessels, the grade of collateralization and therefore
the visibility of the vessel had no influence on revalcularization. In the RCA vessels,
a higher Rentrop grade was seen to have a better chance of revascularization. LAD,
left anterior descending artery; LCX, left circumflex artery; RCA, right coronary
artery.
Fig. 4 Bypass blood flow in mL/min measured in the graft leading to occluded vessels using
transit time flow measurement. In none of the three major vessels, collateralization
and preoperative visibility had a significant influence on bypass blood flow. LAD,
left anterior descending artery; LCX, left circumflex artery; RCA, right coronary
artery.
Continuous variables are expressed as the mean ± standard deviation and were compared
using Student's t-test for normally distributed continuous parameter and Mann–Whitney's U-test for not normally distributed variables. Categorical variables were presented
as percentages and analyzed using the chi-square test or Fisher's exact test. Kolmogorov–Smirnov's
test was performed to analyze normal distribution. The association between Rentrop
grade of collateralization and mean graft flow was analyzed by one-way analysis of
variance (ANOVA) or Kruskal–Wallis' test. Post hoc analysis was performed within the
groups by using the Scheffé's test and Bonferroni's test. Spearman's rank correlation
analysis was used to evaluate the correlation of intraoperatively measured coronary
diameter and Rentrop grade.
A multiple linear regression model was performed and analyzed using an ANOVA. The
p-values of <0.05 are considered as statistically significant.
Results
Patient Characteristics
Of all 938 patients who underwent isolated CABG surgery between 2014 and 2016, 404
patients were identified with an occlusion of at least one coronary artery, resulting
in a CTO prevalence of 43.1%.
In these 404 patients, 621 CTOs were identified, most commonly located in the RCA
territory. CTO was observed in more than one vessel in 102 patients (25.2%). One hundred
twenty-nine (20.8%) patients had an occluded LAD, 131 (21.1%) the LCX, and 259 (41.7%)
the RCA, or one of their major branches was affected by total occlusion ([Fig. 2]).
[Table 1] shows the baseline characteristics of the patients with identified CTOs divided
by Rentrop classification. It shows that patients who had no collaterals were significantly
older and had a significantly reduced creatinine clearance compared with patients
with collaterals. Apart from that, the groups were comparable.
The clinical presentation, the day the patients were admitted to our unit, is shown
in [Table 2]. We saw that patients with poor collateralization had a significantly higher incidence
of ST elevation MI (STEMI) or a non-ST elevation MI. Probably due to that, the number
of patients with a mildly reduced EF was significantly higher in the patients without
collateralization but not the number of patients with a highly reduced EF. Patients
without collateralization were graded as higher risk patients when looking at the
EuroSCORE I and the STS score. However, the patients with no collateralization showed
no significant differences in the SYNTAX Score or the SYNTAX II Score even if these
patients seem to have higher score rating in tendency.
Table 2
Clinical presentation and preoperative risk evaluation
|
Rentrop 0 (n = 23)
|
Rentrop 1 (n = 85)
|
Rentrop 2 (n = 141)
|
Rentrop 3 (n = 155)
|
p-Value
|
|
NYHA (mean ± SD)
|
2.87 ± 0.968
|
2.71 ± 0.721
|
2.63 ± 0.814
|
2.63 ± 0.79
|
0.523
|
|
CCS (mean ± SD)
|
2.78 ± 0.951
|
2.11 ± 1.024
|
1.99 ± 1.119
|
2.12 ± 1.19
|
0.019
|
|
NSTEMI (%)
|
17 (73.91)
|
50 (58.82)
|
68 (48.23)
|
69 (44.52)
|
0.021
|
|
STEMI (%)
|
7 (30.44)
|
12 (14.12)
|
10 (7.09)
|
12 (7.74)
|
0.003
|
|
LVEF >50% (%)
|
6 (26.09)
|
48 (56.47)
|
77 (54.61)
|
88 (56.77)
|
0.048
|
|
LVEF 30–50% (%)
|
15 (65.22)
|
30 (35.29)
|
53 (37.59)
|
52 (33.55)
|
0.033
|
|
LVEF <30% (%)
|
2 (8.7)
|
7 (8.24)
|
11 (7.8)
|
15 (9.68)
|
0.95
|
|
LVEF % (mean ± SD)
|
42.48 ± 10.8
|
48.74 ± 14
|
47.94 ± 13.51
|
47.89 ± 13.79
|
0.267
|
|
Left main stenosis (%)
|
11 (47.82)
|
25 (29.41)
|
39 (27.66)
|
49 (34.77)
|
0.271
|
|
EuroSCORE I
|
7.74 ± 4.59
|
4.85 ± 0.96
|
3.56 ± 0.38
|
3.29 ± 0.26
|
0.002
|
|
SYNTAX score
|
32.24 ± 7.89
|
28.24 ± 8.03
|
28.54 ± 8.26
|
27.92 ± 7.78
|
0.119
|
|
SYNTAX II PCI score
|
42.65 ± 11.95
|
36.45 ± 11.33
|
38.46 ± 12.64
|
37.33 ± 11.65
|
0.138
|
|
SYNTAX II CABG score
|
37.84 ± 8.54
|
32.72 ± 11.29
|
32.68 ± 13.59
|
33.35 ± 12.01
|
0.278
|
|
STS score mortality risk
|
5.51 ± 11.07
|
1.9 ± 3.12
|
2.29 ± 3.79
|
1.98 ± 2.73
|
0.001
|
|
STS score mortality or morbidity risk
|
23.08 ± 20.99
|
14.5 ± 11.26
|
15.82 ± 12.77
|
14.91 ± 11.32
|
0.03
|
Abbreviations: CABG, coronary artery bypass grafting; CCS, Canadian Cardiovascular
Society; LVEF, left ventricular ejection fraction; NYHA, New York Heart Association;
NSTEMI, non-ST elevation myocardial infarction; PCI, percutaneous coronary intervention;
STEMI, ST elevation myocardial infarction; STS, Society of Thoracic Surgery.
Not only did the collateralization have no significant effect on the SYNTAX Score,
but patients were also seen to have no differences in revascularization rates when
looking at the LCA-dependent territories. Only RCA-dependent vessels with no collateralization
had significantly lower chance of being revascularized as seen in [Fig. 3].
The procedural success rate for CR was 82.9% among all CTO patients.
Patients with invisible occluded coronary arteries (Rentrop 0) had a revascularization
rate of 82.6%. It was conspicuous that poor collateralization (Rentrop 1) displayed
the lowest rate of successful CR (68.2%).
In patients with LAD occlusion, the revascularization rate was as high as 96.2%, lower
rates were observed when revascularizing LCX-dependent vessels (85.0%) and the RCA-dependent
territory (78.8%) ([Table 3]).
Table 3
Revascularization rate
|
Rentrop 0 (n = 23)
|
Rentrop 1 (n = 85)
|
Rentrop 2 (n = 141)
|
Rentrop 3 (n = 155)
|
Revascularization rate (%) (n = 404)
|
|
Single LAD occlusion (%) (n = 72)
|
7 (100)
|
13 (92.9)
|
22 (95.7)
|
28 (100)
|
70 (97.22)
|
|
Single RCX occlusion (%) (n = 60)
|
8 (88.9)
|
13 (72.2)
|
17 (94.4)
|
13 (86.7)
|
51 (85)
|
|
Single RCA occlusion (%) (n = 170)
|
0 (0)
|
20 (58.8)
|
56 (83.6)
|
58 (87.9)
|
134 (78.82)
|
|
Two occluded vessels (%) (n = 91)
|
4 (100)
|
11 (68.75)
|
22 (78.6)
|
37 (86)
|
74 (81.32)
|
|
Three occluded vessels (%) (n = 11)
|
0 (0)
|
1 (33.3)
|
3 (75)
|
2 (66.7)
|
6 (54.55)
|
|
Total (%) (n = 404)
|
19 (82.6)
|
58 (68.2)
|
120 (85.1)
|
138 (89)
|
335 (82.92)
|
Abbreviations: LAD, left anterior descending artery; LCX, left circumflex artery;
RCA, right coronary artery.
After being revascularized, the patency of the bypass graft was verified using TTFM.
We saw that blood flow velocity in grafts leading to poorly collateralized vessels
was not significantly different from grafts revascularizing collateralized CTOs. Even
if blood flow to vessels graded as Rentrop 0 seemed to be somewhat lower in tendency
([Table 4] and [Fig. 4]).
Table 4
Mean graft flow measurements according to Rentrop grade
|
Rentrop 0
|
Rentrop 1
|
Rentrop 2
|
Rentrop 3
|
p-Value
|
|
Blood flow LIMA to LAD (mL/min)
|
61.72 ± 45.8
|
69.37 ± 44
|
65.52 ± 35.3
|
59.51 ± 32
|
0.752
|
|
Blood flow venous graft to LCX-dependent vessels (mL/min)
|
47.56 ± 19.9
|
66.67 ± 33.5
|
52.69 ± 36.2
|
64.84 ± 34.8
|
0.219
|
|
Blood flow venous graft to RCA-dependent vessels (mL/min)
|
36 ± 6
|
50.44 ± 35.1
|
61.59 ± 36.1
|
65.03 ± 36.3
|
0.15
|
Abbreviations: LAD, left anterior descending artery; LCX, left circumflex artery;
LIMA, left internal mammary artery; RCA, right coronary artery.
Using the blood flow velocity as a parameter for bypass quality, we computed a multiple
linear regression model. We aimed at identifying parameters that would predict good
flow characteristics in the different coronary artery territories and in the differently
graded CTOs. We found no good predictors for all coronary artery territories. If anything,
one could say that the tallness of a patient has a positive influence on blood flow
velocity in the LIMA graft if anastomosed to an occluded LAD, if graded as Rentrop
1 or 2 ([Table 5]).
Table 5
Multiple linear regression analysis aiming at predictors of graft flow in patients
with CTO and different Rentrop scores
|
Dependent variable: graft flow LIMA to LAD
|
|
Rentrop score
|
Rentrop 0 (n = 18)
|
Rentrop 1 (n = 75)
|
Rentrop 2 (n = 129)
|
Rentrop 3 (n = 139)
|
|
Constant
|
136,905 ± 473,574
|
|
|
−490,456 ± 168,531
|
|
|
−157,188 ± 105,074
|
|
|
15,397 ± 104,391
|
|
|
|
β
|
Std. error
|
Adj. R
2
|
p-Value
|
β
|
Std. error
|
Adj. R
2
|
p-Value
|
β
|
Std. error
|
Adj. R
2
|
p-Value
|
β
|
Std. error
|
Adj. R
2
|
p-Value
|
|
Predictor variable
|
|
|
0.49
|
|
|
|
0.095
|
|
|
|
0.098
|
|
|
|
0.043
|
|
|
Age
|
−6.635
|
2.369
|
|
0.049
|
1.520
|
0.929
|
|
0.107
|
−0.213
|
0.608
|
|
0.727
|
−0.895
|
0.602
|
|
0.140
|
|
Height
|
0.859
|
2.092
|
|
0.702
|
2.184
|
0.793
|
|
0.008
|
0.985
|
0.450
|
|
0.031
|
0.515
|
0.449
|
|
0.254
|
|
BMI
|
2.754
|
2.176
|
|
0.274
|
1.742
|
1.087
|
|
0.114
|
1.633
|
0.832
|
|
0.052
|
−0.021
|
0.728
|
|
0.977
|
|
LVEF
|
0.168
|
1.611
|
|
0.922
|
0.200
|
0.484
|
|
0.681
|
−0.657
|
0.301
|
|
0.031
|
−0.270
|
0.269
|
|
0.318
|
|
EuroSCORE I
|
2.959
|
3.358
|
|
0.428
|
−1.258
|
2.247
|
|
0.578
|
1.149
|
1.524
|
|
0.452
|
2.504
|
1.631
|
|
0.127
|
|
STS mortality risk
|
5.912
|
15.364
|
|
0.720
|
−6.140
|
8.920
|
|
0.494
|
7.525
|
3.803
|
|
0.050
|
−3.827
|
5.469
|
|
0.485
|
|
STS morbidity or mortality risk
|
−2.347
|
3.769
|
|
0.567
|
1.571
|
1.937
|
|
0.420
|
−2.397
|
1.004
|
|
0.019
|
−0.351
|
1.436
|
|
0.807
|
|
SYNTAX score
|
−0.452
|
2.957
|
|
0.886
|
1.631
|
1.318
|
|
0.221
|
1.020
|
1.020
|
|
0.319
|
0.831
|
0.874
|
|
0.344
|
|
SYNTAX II PCI score
|
−2.973
|
4.311
|
|
0.528
|
2.043
|
1.652
|
|
0.221
|
0.972
|
1.006
|
|
0.336
|
0.212
|
1.015
|
|
0.835
|
|
SYNTAX II PCI 4 year mortality
|
0.790
|
3.134
|
|
0.814
|
−0.449
|
0.961
|
|
0.642
|
−0.652
|
0.595
|
|
0.276
|
0.224
|
0.610
|
|
0.714
|
|
SYNTAX II CABG score
|
9.728
|
4.044
|
|
0.074
|
−1.757
|
1.466
|
|
0.235
|
0.509
|
0.858
|
|
0.554
|
1.168
|
1.055
|
|
0.270
|
|
SYNTAX CABG 4 y mortality
|
−3.726
|
3.321
|
|
0.325
|
0.574
|
0.828
|
|
0.491
|
0.132
|
0.492
|
|
0.788
|
−1.002
|
0.795
|
|
0.209
|
Abbreviations: BMI, body mass index; CABG, coronary artery bypass grafting; CTO, chronic
total occlusion; LAD, left anterior descending artery; LVEF, left ventricular ejection
fraction; PCI, percutaneous coronary intervention; STS, Society of Thoracic Surgery.
We also compared the intraoperative measurement of coronary vessel size. This was
routinely done using 1 to 2 mm vessel probes. Correlation analysis demonstrated no
significant correlation between intraoperative measured coronary diameter in CTOs
and Rentrop grade of collateral recipient filling in LAD (r
s = 0.114, p = 0.219) and RCA (r
s = 0.105, p = 0.137) territories. However, evaluation showed a weak positive correlation between
the grade of collateralization and coronary vessel diameter in LCX territory (r
s = 0.295, p = 0.002).
Discussion
Preoperative coronary assessment is crucial for optimal planning of the surgical strategy.
It is seen as one of the most important factors for the patient's future well-being.[18] Many studies have shown that complete coronary revascularization leads to a reduction
in long-term mortality making this the aim of every CABG operation or PCI intervention.[13]
[19]
[20] CTOs increase the complexity of the PCI approach and CR is almost impossible to
achieve if collaterals are scarce or even absent. In CABG surgery, CTOs pose smaller
threat but always leave an uncertainty when planning the operation as one cannot be
sure what size vessels one might find and whether the vessel is worth the graft.[21]
Due to the unknown status of the target vessel and the uncertainty when planning CABG
surgery, one might assume that preoperative not or barely visible occluded coronary
arteries come along with a higher risk of not being revascularizable due to small
vessel size or severe calcification. Unexpectedly, our data showed that patients undergoing
isolated CABG surgery revealed the contrary: Preoperatively assessed invisible or
scarcely visible occluded coronary arteries were seen to have equivalent intraoperatively
measured vessel sizes compared with well visible CTOs. In contrast to LAD- and RCA-dependent
vessels, our analysis only showed a weak correlation between the grade of collateralization
and the vessel size in LCX-dependent arteries. Our study therefore confirms the observation
which revealed no significant target vessel diameter differences between Rentrop gradings.[22] This means that the actual vessel size cannot be predicted by preoperatively evaluated
grade of collateralization and occluded coronary arteries probably will appear as
vessels worth revascularizing even if invisible in the preoperative angiogram.
Furthermore, in the regression analysis, we could not find a good predictor for the
quality of the bypass leading to an occluded vessel. If any, taller patients seem
to have better blood flows on their LIMA grafts if these lead to occluded vessels
graded as Rentrop 1 or 2. We think this is probably caused by the fact that more patients
have more LIMA grafts and therefore might have higher blood flow on the graft. In
general, we do not think this should have an impact on decision making.
The overall CTO prevalence in patients with CAD is estimated to be less than 20%.[2] In our study, 43.1% of the referred patients had an occluded artery, so CTO has
to be regarded as a major reason for referral to surgery. We could also show that
CABG surgery seems to be the superior therapy for achievement of CR in CTO patients.
In our nonselected patient cohort, we completely revascularized 82.6% of the patients
and even achieved revascularization rates of 96.2% if the LAD was occluded. In comparison,
published revascularization rates in selected PCI cohorts usually do not exceed 80%
except for single-experienced centers where at the most 50% of all-comers were treated
achieving a 90% success rate.[23]
[24]
The focus in CTO revascularization—as in all therapies—needs to be on patient benefit.
Unfortunately, this study was not designed to answer this question in total. Especially,
revascularizing vessels seemingly lacking collaterals in the angiography might appear
senseless as these could only perfuse scar tissue and not improve the patients' outcome.
Nevertheless, our findings show that even vessels graded as Rentrop 0 usually have
a revascularizable size and lumen. This makes it hard to believe that these vessels
have no blood flow and leave no viable cardiomyocytes. This feeling is supported by
a nuclear imaging study in which collateral blood flow graded by Rentrop could not
predict myocardial viability or functional improvement.[25] Another study showed that revascularizing not infarct-related CTO vessels increases
LV function in the CTO territory after STEMI.[26]
We tried to support our hypothesis by looking at bypass blood flow as a parameter
for bypass quality. One might think that chronically occluded target vessels lacking
collateralization are more severely diseased downstream of the occlusion than vessels
with retrograde blood flow. This would have an influence on bypass blood flow and
graft patency.[21] We therefore quantified bypass blood flow using TTFM. We were able to show that
occluded vessels lacking collateralization had no significant reduction in blood flow.
We therefore confirm Oshima et al's statement that collateralization—if graded according
to Rentrop—seems to have no influence on bypass blood flow.[21] As low flow in the intraoperative TTFM has been shown to predict early graft failure,
it is our opinion that long-term graft patency due to low flow need not be a concern
in CTOs even if no collateralization is seen in the preoperative angiogram.[21]
Overall, our data show that we should not utterly rely on the preoperative angiogram
when evaluating occluded coronary arteries. Especially in patients with poor collateralization,
we should expect to get more than we see. These vessels were seen to be revascularizable
with a good chance of long-term graft patency. Knowing that, we need to focus on answering
the question of patient benefit and long-term outcome in the future.
