Advances in Interventional Cardiology
Assessment and Impact of the Human Coronary Collateral
Circulation on Myocardial Ischemia and Outcome
Christian Seiler, MD
C ardiovascular disease is the leading cause of death in
industrialized countries and may become the most prev-
alent reason for mortality worldwide.1,2 In the vast majority of
patients, death is the consequence of cerebral or myocardial
infarction, both of which are characterized by acute vessel
occlusion in the context of atherosclerosis.2 In this situation,
outcome depends critically on the extent of infarcted tissue.3
Myocardial infarct size is reflected by the degree of ECG
ST-segment elevation during the acute phase4 and is directly
determined by the duration of coronary occlusion, the isch-
emic area at risk for infarct, the lack of collateral supply to
the jeopardized region, the absence of preconditioning isch-
emic episodes before, and the level of myocardial oxygen
consumption during the acute coronary event.5 The ischemic
area at risk for necrosis is intimately related to alternative
sources of blood supply to the occluded vascular region; this
is to the extent that well-developed coronary anastomoses
originating from a collateral supplying artery can entirely
compensate for the occlusion in the collateral-receiving
vessel, thus rendering its area at risk zero (Figure 1). In
this event, the duration of coronary artery occlusion also
Downloaded from http://ahajournals.org by on April 23, 2021
becomes irrelevant. However, in the event of acute and per-
manent coronary occlusion, myocardium is only salvaged in
the presence of a preformed collateral circulation.6 In this
context, the promotion of well-functioning coronary anas-
tomoses (arteriogenesis) in patients with chronic coronary
artery disease (CAD) before a coronary event is an appealing
therapeutic principle. The selection of suitable candidates
for arteriogenesis should be based on sound assessment of
coronary collateral function.
The purpose of this article is to review the theoretical back-
ground and the different methods for the assessment of col-
lateral function, and to provide an overview of its impact on
myocardial ischemia and outcome.
Theoretical Aspects Relevant to the Assessment
of Coronary Collaterals
The basic physiological and physical principles opera-
tive in the coronary circulation in general also apply to its
anastomoses, that is, the collateral vessels. Oxygen is the
main nutrient for the myocardium, whose demand is deter-
mined by ventricular wall stress, heart rate, and myocardial
Received August 26, 2013; accepted November 11, 2013.
From the Department of Cardiology, University Hospital Bern, Switzerland.
Correspondence to Christian Seiler, MD, Department of Cardiology, University Hospital Bern, CH-3010 Bern, Switzerland. E-mail christian.seiler@insel.ch
(Circ Cardiovasc Interv. 2013;6:719-728.)
© 2013 American Heart Association, Inc.
Circ Cardiovasc Interv is available at http://circinterventions.ahajournals.org DOI: 10.1161/CIRCINTERVENTIONS.113.000555
719
contractility.4 Oxygen extraction from hemoglobin is close
to maximal in myocardial tissue under normal metabolic
conditions and, therefore, changes in oxygen demand are
regulated by altering rates of coronary blood flow (Q in ml/
min). According to Ohm’s law, the perfusion pressure drop
(ΔP in mm Hg) along increasingly smaller tubes can be
described as the product of the vascular resistance (R in dyn
s cm−5) to flow and the flow rate Q: ΔP=R∙Q. In the coronary
circulation, ΔP is the difference between mean aortic perfu-
sion pressure (Pao, mm Hg) and mean central venous pres-
sure (CVP, mm Hg). In the normal coronary circulation, R is
mainly attributable to viscous friction between the blood and
the endothelium covering the wall of small vessels; this can
be appreciated by the fact that the pressure drop in a normal
epicardial coronary tree is equal to only ≈5 mm Hg. Based on
Ohm’s law, pressure drop along the vascular path is further
described by Hagen–Poiseuille’s law, which specifies the
components of vascular geometry contributing to its resis-
tance against flow (R=8 ηl/πr4; with l being vascular seg-
ment length, r being the internal vessel radius, and η is the
blood viscosity). The balance between 2 energy-consuming
factors related to the transport of blood, that is, the cost of
pumping the blood through the circulation as opposed to the
cost of building and maintaining the circulation, defines the
term minimum energy dissipation, the principle of which is
compatible with the structural design of the entire coronary
artery tree including its anastomoses.7 Using various mea-
surement techniques, normal myocardial perfusion under
resting conditions has been documented to be ≈1 mL/min
per gram of tissue.8 Thus numerically (unity between the
flow rate and the regional mass in grams), Q in Ohm’s or
Hagen–Poiseuille’s law can be replaced by regional myo-
cardial mass (M, supplied by blood at any point of interest
in the coronary tree), which is equal to the ischemic area
at risk (AR) for myocardial infarction. AR and M can also
be defined angiographically in terms of summed coronary
artery branch lengths distal to any point in the coronary tree
relative to the entire coronary artery tree length.9 This defini-
tion of AR establishes the relation between collateral vessels
and AR as intuitively outlined above (Figure 1).
In the experimental animal model, the function of arterial
collateral pathways is ideally assessed by perfusion (Q/M)
or conductance measurements (the inverse of R) using the
 720  Circ Cardiovasc Interv  December 2013
instrument of microsphere injections.10 In the clinical setting,
direct myocardial perfusion, flow (Q), or conductance mea-
surements during temporary coronary occlusion is possible
only by quantitative myocardial contrast echocardiography,11
which is, however, technically challenging in the acquisition
of suitable images. Coronary artery occlusion, either tempo-
rary or permanent (as in chronic total occlusion), is required
for unequivocal assessment of collateral supply to a vascu-
lar region of interest (see below). Thus, the question can be
raised, whether obtaining the third parameter of Ohm’s law,
ΔP, which is more robust to determine in the coronary circula-
tion, would yield results, which reflect collateral function reli-
ably and quantitatively. In terms of Ohm’s law, the reference
for collateral function assessment is myocardial perfusion
during coronary artery occlusion (Qoccl/M) relative to normal
myocardial perfusion during vessel patency (Qn/M; suffix n
for normal) for the identical AR or M, that is, a collateral per-
fusion index, CPI (Table 1):
CPI = Q occl / M ÷ Q n / M = Q occl / Q n
(1)
= ( Poccl − CVP ) / R myo  ÷ ( Pn − CVP ) / R myo 

where Poccl is the coronary pressure downstream of the occlu-
sion, CVP is the central venous pressure, Rmyo is the vascu-
lar, microcirculatory resistance of the myocardium within the
AR of interest, and Pn is the mean aortic pressure Pao. Under
the condition of maximal vasodilation using, for example,
adenosine, Rmyo during coronary occlusion and during vessel
patency is minimal and likely similar; from that, it follows that
Downloaded from http://ahajournals.org by on April 23, 2021
CPI ≈ ( Poccl − CVP ) / ( Pao − CVP ) (2)
 pectoris was first described by Heberden in 180213 (Table 2).
Thus, a collateral flow index CFI (mm Hg/mm Hg) can be
defined (Figure 2), which is based purely on cardiac and sys-
temic pressure measurements, and which has been evaluated
clinically (see also below) and found accurate in comparison
with simultaneously obtained echocardiographic myocardial
perfusion measurements12 (Table 1):
CFI = ( Poccl − CVP ) / ( Pao − CVP ) ≈ CPI  (3)
Considering, on the contrary, unequal microvascular resis-
tances Rmyo during vessel occlusion versus patency, it should
be directly obtained:
R myo = (P − CVP ) / Q  (4)
Q = V·A  (5)
where V is mean coronary flow velocity, and A is coronary
cross-sectional area at the site, where V is obtained. If A is
held constant, V goes along with Q. In comparison with Q, V
can be obtained more easily in the human coronary circulation
than Q (Doppler measurements during coronary occlusion
still being challenging). In this context, a coronary collateral
resistance index (CRI) can be obtained on the basis of com-
bined coronary pressure and velocity measurements (Table 1;
see also below):
CRI = ( Poccl − CVP ) / Voccl  ÷ ( Pao − CVP ) / Vnon − occl  (6)

Noninvasive Characterization of the Coronary
Collateral Circulation
In the context of the clinical examination, the only specific
sign hinting at a well-developed coronary collateral circula-
tion is the patient’s information about developing tolerance
to repetitive bouts of myocardial ischemia, that is, relief of
angina pectoris during ongoing or briefly interrupted physical
exercise. This so-called warm-up or walking-through angina
As a pathophysiologic alternative to collateral recruitment on
increased myocardial oxygen demand, it can be explained by
the mechanism of ischemic preconditioning.14 Indirect and
imprecise signs of a functional collateral circulation relate to
the fact that severely narrowed coronary arteries are directly,
although loosely, associated with more extended anastomoses
Figure 1. Normal left ventricular angio-
gram (top middle and right) in a patient
with chronic total occlusion of the proxi-
mal left anterior descending artery occlu-
sion (top left). Angiography of the right
coronary artery (bottom) shows extensive
collateral supply to the naturally occluded
(left) and to the artificially occluded
(right) left anterior descending artery via
apical branch collaterals and via septal
collaterals.
 Seiler   Assessment and Impact of Coronary Collaterals   721

Table 1.  Equations for the Derivation of Quantitative Coronary Collateral Function
Method Equation Measurement Technique Disadvantage
CPI (Qoccl/M)/(Qn/M) Myocardial contrast echocardiography Variable image quality
Positron emission tomography Only in CTO
CFIp (Poccl−CVP)/(Pao−CVP) Pressure sensor guidewire (0.014″) Not in ACS
CFIv Voccl/Vnonoccl Doppler sensor guidewire (0.014″) Voccl difficult to obtain
CRI [(Poccl−CVP)/Voccl]/[(Pao−CVP)/Vnonoccl] Combined pressure/Doppler sensor wire Voccl difficult to obtain
ACS indicates acute coronary syndrome; CFIp, pressure-derived collateral flow index; CFIv, velocity-derived collateral
flow index; CPI, collateral perfusion index; CRI, collateral resistance index; CTO, chronic total (coronary) occlusion; CVP,
central venous pressure; M, regional myocardial mass; Pao, mean aortic pressure; Poccl, mean coronary occlusive pressure;
Qn, normal coronary flow during vessel patency; Qoccl, coronary flow during vessel occlusion; Vnonoccl, coronary flow velocity
during vessel patency; and Voccl, coronary flow velocity during vessel occlusion.

between vascular territories.15,16 In chronic stable CAD, indi-
cators for severe coronary artery stenoses are the degree of
angina pectoris, the level of physical effort at which ECG
signs of ischemia or chest pain occur, and the incidence of
specific ECG signs during exercise. Resting bradycardia,
which is unrelated to betablockade, may indicate collateral
promoting, that is, arteriogenic effects.17
In comparison with chronic CAD, the setting of acute ECG
ST elevation myocardial infarction is more distinctive for the
purpose of noninvasive collateral assessment, because the
acute total coronary occlusion allows the interpretation of the
extent and the degree of ECG ST elevation on the basis of its
contributing factors, area at risk, and collateral supply. The
initial standard 12-lead ECG provides insight into AR, collat-
eral flow, and can estimate subsequent infarct size.18
Downloaded from http://ahajournals.org by on April 23, 2021
noninvasively in this natural occlusion model by different
imaging techniques which, in the case of positron emission
tomography using ammonia as tracer, can even measure abso-
lute tissue perfusion in ml/min gram (Table 1). Because the
invasive examination is needed to confirm CTO in the natural
occlusion model (Figure 1), it is, conversely, essential for an
artificial occlusion model to briefly block the vessel using an
adequately sized angioplasty balloon catheter (Figure 1). The
model of permanent or temporary occlusion of the epicardial
collateral-receiving artery yields the so-called recruitable as
opposed to spontaneously visible collateral flow.
One of the simplest but rather imprecise methods to qualify
collateral function is to register the occurrence of angina pec-
toris shortly before the end of a 1-minute coronary occlusion
(Table 2). However, the predictive value of absent versus pres-

ent chest pain for the distinction between collateral function

Coronary Occlusion Model: Natural
Versus Artificial
As indicated above, invasive cardiac examination is a pre-
requisite for reliable quantitative assessment of the human
coronary collateral circulation, because without a natural
or artificial occlusion of the collateral-receiving artery, the
obtained blood flow reaching the downstream vascular bed
cannot be distinguished by its origin from either the native or
anastomotic path. In patients with CAD, ≈1/4 of all coronary
angiograms show a chronic total (coronary) occlusion (CTO)
of a coronary artery, and only about half of these patients
have viable myocardium in the collateralized area.19 Once an
invasive examination has established the diagnosis of a CTO,
the viable collateralized myocardial region can be examined
sufficient and insufficient to prevent ECG signs of ischemia
is low (sensitivity and specificity to detect sufficient collat-
erals=60%, respectively).20 The severity of chest pain during
myocardial ischemia depends on several factors, such as the
duration of ischemia, the degree of recruitment of collaterals,
previous transmural myocardial infarction, autonomic nerve
dysfunction, psychological and neurobiological characteristics
of the patient, and even the degree of stretching of the coronary
arterial wall by the occluding angioplasty balloon. Irrespective
of simultaneous quantitative collateral measurement, the intra-
coronary ECG at a threshold of ST segment elevation ≥0.1 mV
is widely accepted as a sensitive tool for the detection of isch-
emia.21 Accordingly, an independent dichotomous definition of
coronary collateral vessels insufficient or sufficient to prevent

Figure 2. Pressure-derived coronary

collateral flow index (CFI) measurement
using simultaneous recordings of pha-
sic and mean aortic pressure (Pao, red
line; scale: 200 mm Hg), distal coronary
occlusive pressure (Poccl, black line; scale:
200 mm Hg) and central venous pressure
(CVP, blue line; scale: 50 mm Hg). Addi-
tionally, an intracoronary ECG tracing
is depicted at the bottom, which shows
marked ST-segment elevation during
coronary balloon occlusion.
 722  Circ Cardiovasc Interv  December 2013

Table 2.  Methods for the Assessment of the Coronary Collateral Circulation
Method Precision Advantage Disadvantage Conclusions/Remarks
Warm-up angina Qualitative Obtainable during history taking Low prevalence not reflecting Specific for sufficient collaterals
prevalence of sufficient collaterals
Angina at occlusion Qualitative Easy and costless; no technical Large interindividual and likely Helpful for crude risk stratification
equipment needed intraindividual variability
ECG during occlusion Qualitative Easy and costless; more Large interindividual variability; Helpful for crude risk stratification
accurate than AP variable with artery examined
Regional LV function Qualitative With CTO and normal LV Not applicable in stenotic arteries Applicable in CTOs, otherwise too
function no further assessment without 2nd catheter or elaborate for limited information
necessary simultaneous echo
Angiography Qualitative With natural coronary occlusion Otherwise too elaborate for limited Appropriate only for natural occlusions
easy and costless to perform information (2nd catheter) and crude risk stratification
Washout angiography Semiquantitative Easy and costless to perform; Only semiquantitative Applicable in all coronary arteries;
assesses ipsilateral and more accurate risk stratification
contralateral collaterals; no
second catheter needed
Pressure (CFIp) and Quantitative Feasible in most arteries; robust Challenging in the acquisition of Current gold standard; appropriate for
Doppler sensor (CFIv) pressure measurements Doppler signals during occlusion risk stratification and clinical studies
measurements Poccl not applicable in ACS
Collateral perfusion Quantitative Noninvasive assessment Complex procedure; variable image Appropriate for risk stratification and
measurement (CPI) with CTO quality; extensive postprocessing; clinical studies
operator-dependent
ACS indicates acute coronary syndrome; AP, angina pectoris; CFIp, pressure-derived collateral flow index; CFIv, velocity-derived collateral flow index; CPI, collateral
perfusion index; CTO, chronic total (coronary) occlusion; and LV, left ventricle.

myocardial ischemia of a briefly occluded vascular area is this requires double coronary intubation. In the presence of a
given by the presence (Figure 2) or absence (Figure 3) of intra- CTO, predominant angiographic collateral pathways run via
coronary ECG ST segment elevation ≥0.1 mV. The intracoro- septal collaterals in >2/5 of patients, close to 1/5 via distal
Downloaded from http://ahajournals.org by on April 23, 2021

nary ECG ST-segment shift (ie, elevation or depression) during
a 1-minute coronary occlusion is also quantifiable and, as such,
independently predictive of all-cause mortality.22
Even with several CTOs present, there may be an entirely
normal systolic left ventricle (LV) function attributable to a
well-developed collateral circulation. However, a substantial
number of patients with CTO reveal various degrees of sys-
tolic LV dysfunction. The recovery of impaired systolic LV
function after revascularization of a CTO seems not to depend
on the quality of collateral function,23 suggesting that collat-
eral development does not depend on the presence of viable
myocardium. In patients with CAD without CTO and viable
myocardium, simultaneous assessment of regional LV function
using transthoracic tissue Doppler imaging and invasive collat-
eral function has shown an association between systolic as well
as diastolic LV function and collateral function24 (Table 2).
Angiographic Collateral Assessment
The most widely used invasive method for assessing the coro-
nary collateral circulation is contrast angiography because of
its availability and the superior imaging quality concerning
the often small-sized collateral vessels (Figure 1). As opposed
to the coronary patency method used during single vessel
intubation with assessment of spontaneously visible collat-
eral vessels, the coronary occlusion model images recruitable
collaterals. With the exception of the 1/4 patients with CAD
with CTO encountered during coronary angiography and of
patients undergoing primary percutaneous coronary interven-
tion (PCI) in acute myocardial infarction, angiographic collat-
erals are usually not assessed by the occlusion model, because
branch collaterals, and ≈1/3 via atrial collaterals with proxi-
mal takeoff.25 The normal human heart and that affected by
CAD contain numerous anastomotic vessels ranging between
40 and 200 μm.26 Hence, the size of the majority of these
vessels is below the spatial resolution even of analog angio-
graphic imaging chains.
The conventional angiographic grading system is that
originally described by Rentrop et al,27 who distinguished 4
degrees of collateral recipient artery filling by radiographic
contrast medium: grade 0=no collaterals; grade 1=side branch
filling of the recipient artery without filling of the main epicar-
dial artery; grade 2=partial filling of the main epicardial recip-
ient artery; grade 3=complete filling of the main epicardial
recipient artery (Figure 1). Of note, occlusion of the collateral-
receiving artery clearly augments the sensitivity of detecting
collateral vessels using angiography but renders the method
of artificial occlusion technically more demanding, because of
the necessity of double coronary ostial intubation (Figure 1).
A semiquantitative angiographic method for collateral
assessment in patients undergoing CTO recanalization has
been introduced25 using the recipient filling grade according
to Rentrop et al,27 their predominant anatomic location, and
a new grading of collateral connections (CC grade 0=no con-
tinuous connection between collateral supplying and receiv-
ing vessel, in 14%; CC1=threadlike continuous connection, in
51%; CC2=side-branch like connection, in 35%). Similarly,
but without the difficulty of identifying collateral versus
recipient vessels, the time to clearance of radiographic con-
trast medium (washout) trapped distal to a balloon-occluded
collateral-receiving artery can be determined. Washout at a
 Seiler   Assessment and Impact of Coronary Collaterals   723

Figure 3. Pressure-derived coronary collateral flow index (CFI) measurement in the same patient as shown in Figure 2 with simultaneous
pressure recordings (mean aortic pressure [Pao], mean coronary occlusive pressure [Poccl], central venous pressure [CVP]); recording of
Poccl downstream of the proximally occluded left anterior descending artery (LAD; left), and downstream of the proximally occluded right
coronary artery after percutaneous coronary intervention of the LAD occlusion (right). CFI in both coronary arteries is equal to ≈0.8 and
sufficient to prevent signs of myocardial ischemia as indicated by the intracoronary ECG (lowest lead).

threshold of 11 heartbeats accurately distinguishes between fractional blood flow by coronary pressure measurements;
Downloaded from http://ahajournals.org by on April 23, 2021

collateral supply sufficient or insufficient to prevent myocar-
dial ischemia during a brief occlusion (sensitivity of 88%,
specificity of 81%28; Table 2).
Quantitative Coronary Pressure and Doppler
Sensor Measurements
In addition to their principle angiographic results, Rentrop
et al27 first described an angioplasty balloon occlusion model
using the ratio between distal coronary occlusive or wedge
pressure (Poccl) and aortic pressure (Pao), which correlates with
angiographic collateral score groups in patients with 1-vessel
CAD (Table 2).29 In 1987, Meier et al30 found that a Poccl ≥30
mm Hg accurately predicted the presence of spontaneously
visible or recruitable collaterals. However, Poccl is not depen-
dent only on the amount of collateral flow to the occluded
vascular region. Determinants of Poccl are the driving pressure
across collateral pathways (ie, the pressure difference between
the coronary pressure in the collateral supplying and receiv-
ing artery), the venous back pressure (central venous or right
atrial pressure), but also extravascular pressure related to com-
pression of intramural vessels by cardiac contraction and to
transmission of diastolic LV pressure to the epicardial circu-
lation.31 In this context, in that of microembolization of the
collateral-dependent vascular area, Poccl should not be used for
collateral assessment in the acute coronary syndrome because
collateral function is substantially overestimated. In the setting
of chronic CAD, LV filling pressure has, however, no influence
on Poccl unless it exceeds values of 30 mm Hg.
In 1993, Pijls et al32 provided the experimental basis for
the determination of maximum coronary and myocardial
importantly, the contribution to myocardial blood flow by
collateral pathways was assumed negligible. In this study,
microvascular resistance during pharmacologically induced
hyperemia was postulated to be constant and minimal and,
thus, no longer dependent on the degree of epicardial stenoses.
As a consequence, coronary pressure should directly reflect
coronary flow, the former of which is more easily obtainable
during invasive examination than the latter. Therefore, myo-
cardial flow Q distal to a stenosis (the sum of flow through
the stenotic vessel plus collateral flow as assumed to be
zero) relative to myocardial flow without the stenosis QN (Q/
QN=fractional flow reserve, FFR) could be calculated on the
basis of distal coronary pressure and aortic pressure both after
subtraction of central venous backpressure (CVP). The model
was tested in dogs, whereby Doppler-derived measurements
of stenotic coronary flow Qs to normal flow QN was directly
compared with (Pd-CVP)/(Pao-CVP). However, the situation
of a complete coronary occlusion (Qs=0, ie, the setting where
Pd=Poccl) was not tested.32 In fact, direct and simultaneous eval-
uation of the coronary parameters quantitative for collateral
function ([Poccl−CVP]/[Pao−CVP]; Figures 2 and 3) in compar-
ison with occlusive coronary flow velocity was not performed
until 1998: coronary pressure- and Doppler-derived ratios
indicative of collateral function during PCI were compared
in patients with CAD and termed pressure-derived (CFIp) and
velocity-derived collateral flow index, respectively (CFIv33;
see also Table 1). Figure 2 shows an example of CFIp values
insufficient to prevent ECG signs of myocardial ischemia (ST
segment shift ≥0.1 mV) during coronary occlusion. Close to
1800 CFIp measurements using the above intracoronary ECG
 724  Circ Cardiovasc Interv  December 2013
definition of myocardial ischemia (ST elevation ≥0.1 mV)
provide the best cutoff of 0.217 for the most accurate detection
of sufficient and insufficient collaterals (76% sensitivity and
76% specificity),20 which is in close agreement with a study
among patients with acute myocardial infarction undergoing
single photon emission tomography before primary PCI.34
Potential Pitfalls, Limitations, and Risks
Mean CVP should be obtained systematically as temporal
average during several respiratory cycles. During pressure
recordings, the patient should be asked to breathe normally
and not to speak to maintain physiological CVP variations. As
further technical aspects, Pd or Poccl pressure shifts attributable
to leakage of electric current and artificial systolic pressure
peaks in relation to looping of the pressure guidewire have to
be considered. Both problems occur more often during pro-
longed use and technically demanding maneuvering of the
wire. Doppler- or velocity-derived collateral assessment by
Doppler-tipped guide wires is much less robust than pressure-
derived CFI measurement (Tables 1 and 2). This is mainly
because of difficulties of differentiating low occlusive coro-
nary flow velocity signals from vascular wall motion artifacts,
respectively, to time-consuming efforts of wire repositioning
to obtain true flow velocity signals.
In ≈2/3 of patients with chronic CAD, coronary occlusion
causes myocardial ischemia,16 which is considered a strong
hyperemic stimulus. An additional pharmacological hyper-
emic stimulus (eg, by intravenous adenosine) most likely does
not induce further reduction of microvascular resistance in this
Downloaded from http://ahajournals.org by on April 23, 2021
population, that is, it does not alter CFI. However, among indi-
viduals not revealing signs of ischemia during occlusion, phar-
macological vasodilation may further decrease collateral and
peripheral vascular resistance and increase CFI. Alternatively,
microvascular resistance in the collateral supplying region may
predominantly decline, and collateral flow may be redirected
away from the collateral-receiving area (ie, collateral steal).35
The maximum vasodilatory stimulus is specifically impor-
tant when assessing the natural occlusion model of a CTO.
In this situation, we cannot assume the presence of spontane-
ous maximum hyperemia because the occlusion is permanent.
Vasodilation using systemic adenosine in these patients pro-
vides a wide variation of responses of the collateral flow and
pressure recordings, which are not unidirectional.36
Angioplasty balloon occlusion of a normal coronary
artery for the purpose of CFI measurement may pose a risk
for endothelial injury and development of a de novo stenosis.
Aside from the shortness of a 1-minute vessel occlusion, the
principal feature of our protocol with regard to preventing
vessel injury is the use of a low balloon inflation pressure
just sufficient to occlude the artery. This minimal occlu-
sion pressure is reached slowly, and imminent occlusion
is sensed using the start of pressure decline obtained dis-
tal to the balloon and not primarily angiographic detection
of occlusion, which only follows later. In angiographically
normal arteries, an analysis of 426 measurements revealed
a dissection in 1 vessel (1/426=0.2%), subsequently treated
by stent implantation.37 In 35% of all vessels investigated
(n=150; mean follow-up of 10 months), angiography was
repeated most often because of planned exams. In 2 out of
150 patients (=1.3%) both having progressive CAD, a new
stenosis at the site of balloon occlusion occurred 14 and 72
months after the initial occlusion, respectively.37
Quantitative Collateral Perfusion Measurements
A direct verification of CFIp versus the reference of myocar-
dial blood supply has been performed recently: Myocardial
perfusion (ml/min per gram), defined as blood flow Q into
a region relative to its mass M, can be obtained by positron
emission tomography and lately by myocardial contrast echo-
cardiography (MCE; see below for description of the tech-
nique; Table 1).11 Direct comparison of CFIp and absolute
myocardial perfusion to a briefly and artificially occluded vas-
cular region requires a bedside quantitative method for blood
flow measurements, a condition fulfilled by MCE. Two human
studies with coronary occlusion to avoid concomitant contrast
flow via the native vessel, compared MCE and invasive col-
lateral assessment.38,39 In patients with recent acute myocar-
dial infarction, angiographically visible collaterals correlated
poorly with the size of the collateralized area as well as nor-
malized ultrasound contrast agent transit rates.38 This is not
unexpected because angiographic collateral vessels develop
only late in about half the patients after infarction. In patients
with stable CAD undergoing PCI, CFIp has been shown to cor-
relate modestly with peak signal intensity of ultrasound con-
trast agent transit curves but not with contrast transit rates.39
Based on the MCE technique of ultrasound contrast agent
destruction by high mechanical index with subsequent obser-
vation of contrast refill (expressed by the volume exchange
rate β) within a vascularized myocardial region of interest
(providing the parameter of relative myocardial blood vol-
ume, rBV), absolute collateral myocardial perfusion or collat-
eral myocardial blood flow has been obtained during coronary
angioplasty balloon occlusion in 30 patients undergoing PCI12
(Table 2). Myocardial blood flow has been calculated accord-
ing to the continuity equation as the product of β and rBV
divided by myocardial tissue density. The precision of this
MCE method has been previously documented in comparison
with a perfusion phantom model, to positron emission tomog-
raphy and to invasive coronary flow velocity measurements.11
In the context of our study with side-by-side comparison of 2
different quantitative methods for collateral assessment,12 it is
reasonable to state that CFIp measurements accurately reflect
collateral relative to normal antegrade flow in humans with
chronic stable CAD, even in the low range. The overestima-
tion of MCE-derived collateral perfusion index (collateral
relative to normal myocardial perfusion; CPI) by pressure-
derived CFI is much less than that of Doppler-derived CFI,33
and it amounts to ≈2% to 7% depending on whether the CFI-
intercept or the standard error of estimate of the CFIp−CPI
relation is considered. However, MCE-derived collateral
assessment may be limited by insufficient image quality, and
an elaborate postprocessing image analysis (Table 2).
Impact of the Collateral Circulation on
Myocardial Ischemia
As a surrogate for the prognostically relevant infarct size, stud-
ies on myocardial salvage and on myocardial injury attribut-
able to PCI have used the magnitude of intracoronary ECG
 Seiler   Assessment and Impact of Coronary Collaterals   725

Because heart rate reduction has been consistently found more

i.c. ECG ST segment shift (mV) 3
2
1 interval could be used as markers for the degree of ischemia.
0 mulated that QT prolongation or time variability are markers
0.0 0.2 0.4
Collateral flow index, CFIp
Figure 4. Association between collateral flow index (CFI, x axis)
and intracoronary (i.c.) ECG ST-segment shift (y axis). y=1.0–
0.55x; n=809, r2=0.129, P<0.0001. The red broken lines indicate
the ECG ST shift of 0.1 mV above which ischemia is defined, and
the (best) CFI threshold of 0.217 for the distinction between well
and poorly functioning collaterals.
ST-segment shift during or after coronary balloon occlusion.4,40
Intracoronary ECG recording from the angioplasty catheter
guidewire is more sensitive and reliable in detecting regional
myocardial ischemia during balloon inflation than standard
ECG.21 As outlined above, the degree of myocardial ischemia
during coronary occlusion is not only dependent on collateral
supply (Figures 2 and 3), but also on factors such as the occlu-
sion duration, the size of the ischemic area, the occurrence of
preconditioning episodes of ischemia, and the actual level of
oxygen consumption, which itself is determined by heart rate,
blood pressure, and myocardial contractility.5 Thus, the relation
between the degree of ischemia during occlusion and collateral
Downloaded from http://ahajournals.org by on April 23, 2021
prevalent during ischemia of the right than the left coronary
territory,41 the abovementioned determinants low heart rate and
right coronary artery territory cannot be called independent
parameters in a biological sense. Alternatively to intracoro-
nary ECG ST-segment shift, other ECG parameters such as QT
QT interval would be reasonable because evidence has accu-
0.6 0.8 1.0
or triggers of myocardial electric instability and sudden cardiac
death.41,42 A study among 150 patients with stable CAD under-
going a standardized 1-minute coronary balloon occlusion with
simultaneous intracoronary ECG recording has documented a
QT prolongation during ischemia in the left, but not the right
coronary territory, whereby it has been inversely associated to
CFI,41 thus indicating a protective effect of the collateral circu-
lation not only against infarct size expansion, but also against
ischemia-related risk of sudden cardiac death. The threshold
of arterial blood supply in absolute terms has been determined
using quantitative myocardial contrast echocardiography.43
Using this method, the distinction between patients with and
without intracoronary ECG signs of myocardial ischemia (0.1
mV cutoff) during 1 minute of coronary occlusion has been
found to be most accurate at a threshold of 0.374 mL/min
per gram of perfusion.43
Impact of the Collateral Circulation on Outcome
There is a hierarchy of end points defining clinical outcome,

supply can be expected to be inverse but not absolutely tight. In
this context, a recent study on the determinants of quantitatively
assessed intracoronary ECG ST-segment shift during a 1-min-
ute coronary occlusion has been performed in 765 patients
with stable CAD undergoing simultaneous CFI measure-
ment.22 Absence of ECG ST-segment shift (<0.1 mV) during
coronary occlusion has been found to be independently related
to a well-functioning collateral supply to the occluded region
(ie, CFI≥0.217; Figure 4), to a low heart rate during occlusion,
to the right coronary artery territory as the ischemic area, and
to the absence of arterial hypertension in the patient’s history.22
which consists of all-cause mortality, cardiac mortality, car-
diovascular mortality, myocardial infarction, unstable angina
pectoris, repeat revascularization, and the combination of
those events. In addition, surrogate end points used in studies
on the prognostic relevance of collaterals are infarct size, LV
aneurysm formation, LV systolic function, and myocardial via-
bility.6 The beneficial effect of well-developed collaterals on
all surrogate end points has been recognized.44–48 To influence
the pathophysiologic prognosticator of outcome beneficially,
infarct size, collaterals need to be preformed for preventing
LV remodeling and aneurysm formation, whereas they may

Good CCC Poor CCC

Study Total Deaths Total Deaths RR 95%-CI

Helfant 1971 61 6 58 10 0.57 [0.22; 1.47]

Williams 1976 6 0 14 8 0.13 [0.01; 1.95]
Nestico 1985 183 5 176 5 0.96 [0.28; 3.26]
Hansen 1989 67 32 29 19 0.73 [0.51; 1.05]
Perez-Castellano 1998 65 5 115 26 0.34 [0.14; 0.84]
Nicolau 1999 59 16 363 54 1.82 [1.12; 2.96] Figure 5. Impact of the coronary col-
Antioniucci 2002 264 11 900 80 0.47 [0.25; 0.87] lateral circulation (CCC) on survival
Monteiro 2003 35 4 35 6 0.67 [0.21; 2.16] related to all-cause mortality as analyzed
Meier 2007 226 25 586 170 0.38 [0.26; 0.56] in a meta-analysis including 12 studies.
Regieli 2009 263 2 616 4 1.17 [0.22; 6.35] Reproduced from Meier et al51 with per-
Desch 2009 69 2 166 13 0.37 [0.09; 1.60] mission of the European Society of Cardi-
Steg 2010 1922 155 251 28 0.72 [0.49; 1.06] ology. Copyright ©2012, Meier et al.

Random effects model 3220 3309 0.64 [0.45; 0.91]

Heterogeneity: I-squared=66.5%, tau-squared=0.2097, p=0.0006

0.2 0.5 1 2 5
Lower Mortality Higher Mortality
With Good CCC
 726  Circ Cardiovasc Interv  December 2013
develop only after myocardial infarction (see above). During
the first 3 to 6 hours of acute coronary syndrome, angiograph-
ically well visible collateral vessels are present in ≈50%, a
number increasing to ≈100% in the presence of continuing
vascular occlusion.49 Late angiographic appearance signifies
no prognostic benefit because collaterals supplying necrotic
myocardium do not salvage it. In this context, the impact of
the collateral circulation on outcome has to be interpreted dif-
ferently depending on whether the focus is on chronic stable
or on various phases of acute CAD (see below).
The association between the angiographic presence of col-
lateral vessels and medium-term outcome in patients with
chronic CAD has been investigated as early as 1971. The risk
to die from any cause during a follow-up of ≈2 years in that
study by Helfant et al50 was reduced to almost half in the pres-
ence versus the absence of angiographic collaterals. However,
the 95% confidence interval of the relative risk (0.22–1.47)
found in that study rendered the survival benefit of patients
with good collateral supply uncertain and exemplified the
triple limitation of many subsequent studies on the same
topic: low number of recruited patients with a brief observa-
tion time and use of a blunt method for collateral assessment,
that is, coronary angiography. The range of patients included
in 13 studies on the prognostic effect of the coronary collat-
eral circulation is equal to 22 to 2173 (mean 866), the range
of follow-up duration is between 30 days and 16 years, and
the majority of these studies have used angiographic grading
of spontaneously visible collaterals as the method for assess-
ment.20,51 The investigation of Antoniucci et al52 and that of
Downloaded from http://ahajournals.org by on April 23, 2021
Steg et al53 illustrate that the use of a vague method for col-
lateral assessment can be compensated for by including more
patients in a setting, in which mortality is genuinely higher
than in the population with chronic stable CAD (2%/yr):
acute, respectively, subacute myocardial infarction. The rela-
tive risk for mortality from any cause in these studies among
All-Cause Deaths (n=234)
1
.9
Cumulative survival rate
.8
.7
.6 • CFI≥0.25 (n=286)
X CFI<0.25 (n=895)
Mean follow-up =7.3±4.3 years
.5
0 2 4 6 8 10
Years of follow-up
N at risk
CFI≥0.25 286 277 260 251 248 152
CFI<0.25 895 866 810 775 750 460
Figure 6. Cumulative survival rates related to all-cause mortal-
ity in patients with low and with high collateral flow index (CFI).
Reproduced from Seiler et al20 with permission from BMJ Pub-
lishing Group Ltd. Copyright ©2013, BMJ Publishing Group Ltd.
patients with spontaneously visible collateral arteries has been
reduced to 0.47 (95% confidence interval 0.25–0.87) in the
former and to 0.72 (95% confidence interval 0.49–1.06) in the
latter investigation, respectively (Figure 5).50,52–62 The fact that
in the study by Steg et al53 angiographic collaterals were only
marginally beneficial on all-cause mortality is directly related
to the inclusion of about half the patients in this subacute
infarction population with delayed arteriogenesis occurring
only after the myocardial-salvaging window of the 1st 3 to 6
hours after symptom onset.
The main result of a recently published study among 1181
patients with chronic CAD has been that a growing collateral
function as quantitatively assessed by CFI is an independent
beneficial prognosticator.20 The relative risk to die from any
cause in patients with high CFI during this 16-year follow-up
study (Figure 6) was reduced to 0.206 (95% confidence inter-
val 0.067–0.637; P=0.0061).20 This is consistent with the prin-
cipal finding of a recent meta-analysis documenting a relative
risk reduction for all-cause mortality to 0.64 (95% confidence
interval 0.45–0.91) among patients with a well-developed
coronary collateral circulation.51
Conclusions
In patients with CAD, a well-developed coronary collateral
circulation is related to reduction of infarct size, LV dys-
function, and all-cause mortality. Concerning future clinical
research, accurate and quantitative assessment methods are
needed for a reliable distinction between patients with col-
lateral function sufficient or insufficient to prevent ischemia
during coronary occlusion. Invasive methods are inevitable
for collateral assessment with pressure-derived CFI measure-
ments being the current gold standard.
Sources of Funding
This work was supported by a grant from the Swiss National Science
Foundation (to C. Seiler, No. 32003B-124848) and by the Swiss
Heart Foundation.
Disclosures
None.
References
1. Lopez, Mathers C, Ezzati M, Jamison D, Murray C. Global and regional
burden of disease and risk factors, 2001: systematic analysis of popula-
tion health data. Lancet. 2006;367:1747–1757.
2. World Health Organization G. Global atlas on cardiovascular disease
p=0.0057 prevention and control. 2011.
3. Sobel BE, Bresnahan GF, Shell WE, Yoder RD. Estimation of infarct size
in man and its relation to prognosis. Circulation. 1972;46:640–647.
4. Maroko PR, Kjekshus JK, Sobel BE, Watanabe T, Covell JW, Ross J
Jr, Braunwald E. Factors influencing infarct size following experimental
coronary artery occlusions. Circulation. 1971;43:67–82.
5. Schaper W, Frenzel H, Hort W. Experimental coronary artery occlusion.
I. Measurement of infarct size. Basic Res Cardiol. 1979;74:46–53.
12 14 16 6. Seiler C. Collateral Circulation of the Heart. London: Springer; 2009.
7. Seiler C, Kirkeeide RL, Gould KL. Basic structure-function rela-
tions of the epicardial coronary vascular tree. Basis of quantitative
87 49 0 coronary arteriography for diffuse coronary artery disease. Circulation.
252 151 0
1992;85:1987–2003.
8. Bergmann SR, Herrero P, Markham J, Weinheimer CJ, Walsh MN.
Noninvasive quantitation of myocardial blood flow in human subjects
with oxygen-15-labeled water and positron emission tomography. J Am
Coll Cardiol. 1989;14:639–652.
 Seiler   Assessment and Impact of Coronary Collaterals   727
9. Seiler C, Kirkeeide RL, Gould KL. Measurement from arteriograms of re-
gional myocardial bed size distal to any point in the coronary vascular tree
for assessing anatomic area at risk. J Am Coll Cardiol. 1993;21:783–797.
10. Hoefer IE, van Royen N, Buschmann IR, Piek JJ, Schaper W. Time
course of arteriogenesis following femoral artery occlusion in the rabbit.
Cardiovasc Res. 2001;49:609–617.
11. Vogel R, Indermühle A, Reinhardt J, Meier P, Siegrist PT, Namdar M,
Kaufmann PA, Seiler C. The quantification of absolute myocardial perfu-
sion in humans by contrast echocardiography: algorithm and validation.
J Am Coll Cardiol. 2005;45:754–762.
12. Vogel R, Zbinden R, Indermühle A, Windecker S, Meier B, Seiler C.
Collateral-flow measurements in humans by myocardial contrast echo-
cardiography: validation of coronary pressure-derived collateral-flow as-
sessment. Eur Heart J. 2006;27:157–165.
13. Heberden W. Commentaries on the history and cure of diseases. In: Wilius
F, Kays T, eds. Classics of Cardiology. New York: Dover; 1961:220–224.
14. Williams RP, Manou-Stathopoulou V, Redwood SR, Marber MS.
‘Warm-up angina’: harnessing the benefits of exercise and myocardial
ischaemia. Heart. June 18, 2013. doi: 10.1136/heartjnl-2013-304187.
15. Piek JJ, Koolen JJ, Hoedemaker G, David GK, Visser CA, Dunning AJ.
Severity of single-vessel coronary arterial stenosis and duration of an-
gina as determinants of recruitable collateral vessels during balloon oc-
clusion. Am J Cardiol. 1991;67:13–17.
16. Pohl T, Seiler C, Billinger M, Herren E, Wustmann K, Mehta H,
Windecker S, Eberli FR, Meier B. Frequency distribution of collateral
flow and factors influencing collateral channel development. Functional
collateral channel measurement in 450 patients with coronary artery dis-
ease. J Am Coll Cardiol. 2001;38:1872–1878.
17. Lamping KG, Zheng W, Xing D, Christensen LP, Martins J, Tomanek RJ.
Bradycardia stimulates vascular growth during gradual coronary occlu-
sion. Arterioscler Thromb Vasc Biol. 2005;25:2122–2127.
18. Christian TF, Gibbons RJ, Clements IP, Berger PB, Selvester RH,
Wagner GS. Estimates of myocardium at risk and collateral flow in acute
myocardial infarction using electrocardiographic indexes with com-
parison to radionuclide and angiographic measures. J Am Coll Cardiol.
1995;26:388–393.
19. Fefer P, Knudtson ML, Cheema AN, Galbraith PD, Osherov AB,
Downloaded from http://ahajournals.org by on April 23, 2021
Yalonetsky S, Gannot S, Samuel M, Weisbrod M, Bierstone D, Sparkes
JD, Wright GA, Strauss BH. Current perspectives on coronary chronic
total occlusions: the Canadian Multicenter Chronic Total Occlusions
Registry. J Am Coll Cardiol. 2012;59:991–997.
20. Seiler C, Engler R, Berner L, Stoller M, Meier P, Steck H, Traupe T.
Prognostic relevance of coronary collateral function: confounded or
causal relationship? Heart. 2013;99:1408–1414.
21. Friedman PL, Shook TL, Kirshenbaum JM, Selwyn AP, Ganz P. Value
of the intracoronary electrocardiogram to monitor myocardial ischemia
during percutaneous transluminal coronary angioplasty. Circulation.
1986;74:330–339.
22. de Marchi SF, Streuli S, Haefeli P, Gloekler S, Traupe T, Warncke C,
Rimoldi SF, Stortecky S, Steck H, Seiler C. Determinants of prognosti-
cally relevant intracoronary electrocardiogram ST-segment shift during
coronary balloon occlusion. Am J Cardiol. 2012;110:1234–1239.
23. Werner GS, Surber R, Kuethe F, Emig U, Schwarz G, Bahrmann P,
Figulla HR. Collaterals and the recovery of left ventricular function
after recanalization of a chronic total coronary occlusion. Am Heart J.
2005;149:129–137.
24. Seiler C, Pohl T, Lipp E, Hutter D, Meier B. Regional left ventricular
function during transient coronary occlusion: relation with coronary col-
lateral flow. Heart. 2002;88:35–42.
25. Werner GS, Ferrari M, Heinke S, Kuethe F, Surber R, Richartz BM,
Figulla HR. Angiographic assessment of collateral connections in com-
parison with invasively determined collateral function in chronic coro-
nary occlusions. Circulation. 2003;107:1972–1977.
26. Fulton WF. Arterial anastomoses in the coronary circulation. II.
Distribution, enumeration and measurement of coronary arterial anasto-
moses in health and disease. Scott Med J. 1963;8:466–474.
27. Rentrop KP, Cohen M, Blanke H, Phillips RA. Changes in collateral chan-
nel filling immediately after controlled coronary artery occlusion by an an-
gioplasty balloon in human subjects. J Am Coll Cardiol. 1985;5:587–592.
28. Seiler C, Billinger M, Fleisch M, Meier B. Washout collaterometry: a
new method of assessing collaterals using angiographic contrast clear-
ance during coronary occlusion. Heart. 2001;86:540–546.
29. Probst P, Zangl W, Pachinger O. Relation of coronary arterial occlusion
pressure during percutaneous transluminal coronary angioplasty to pres-
ence of collaterals. Am J Cardiol. 1985;55:1264–1269.
30. Meier B, Luethy P, Finci L, Steffenino GD, Rutishauser W. Coronary
wedge pressure in relation to spontaneously visible and recruitable col-
laterals. Circulation. 1987;75:906–913.
31. de Marchi SF, Oswald P, Windecker S, Meier B, Seiler C. Reciprocal
relationship between left ventricular filling pressure and the recruitable
human coronary collateral circulation. Eur Heart J. 2005;26:558–566.
32. Pijls NHJ, van Son JAM, Kirkeeide RL, de Bruyne B, Gould KL.
Experimental basis of determining maximum coronary, myocardial, and
collateral blood flow by pressure measurements for assessing functional
stenosis severity before and after percutaneous coronary angioplasty.
Circulation. 1993;86:1354–1367.
33. Seiler C, Fleisch M, Garachemani A, Meier B. Coronary collateral quanti-
tation in patients with coronary artery disease using intravascular flow ve-
locity or pressure measurements. J Am Coll Cardiol. 1998;32:1272–1279.
34. Christian TF, Berger PB, O’Connor MK, Hodge DO, Gibbons RJ.
Threshold values for preserved viability with a noninvasive measurement
of collateral blood flow during acute myocardial infarction treated by
direct coronary angioplasty. Circulation. 1999;100:2392–2395.
35. Seiler C, Fleisch M, Meier B. Direct intracoronary evidence of collateral
steal in humans. Circulation. 1997;96:4261–4267.
36. Werner GS, Fritzenwanger M, Prochnau D, Schwarz G, Ferrari M,
Aarnoudse W, Pijls NH, Figulla HR. Determinants of coronary steal in
chronic total coronary occlusions donor artery, collateral, and microvas-
cular resistance. J Am Coll Cardiol. 2006;48:51–58.
37. Gloekler S, Traupe T, Meier P, Steck H, de Marchi SF, Seiler C. Safety of
diagnostic balloon occlusion in normal coronary arteries. Am J Cardiol.
2010;105:1716–1722.
38. Sabia PJ, Powers ER, Jayaweera AR, Ragosta M, Kaul S. Functional
significance of collateral blood flow in patients with recent acute myo-
cardial infarction. A study using myocardial contrast echocardiography.
Circulation. 1992;85:2080–2089.
39. de Marchi SF, Schwerzmann M, Fleisch M, Billinger M, Meier B, Seiler
C. Quantitative contrast echocardiographic assessment of collateral de-
rived myocardial perfusion during elective coronary angioplasty. Heart.
2001;86:324–329.
40. Balian V, Galli M, Marcassa C, Cecchin G, Child M, Barlocco F, Petrucci
E, Filippini G, Michi R, Onofri M. Intracoronary ST-segment shift soon
after elective percutaneous coronary intervention accurately predicts
periprocedural myocardial injury. Circulation. 2006;114:1948–1954.
41. Meier P, Gloekler S, de Marchi SF, Zbinden R, Delacrétaz E, Seiler C.
An indicator of sudden cardiac death during brief coronary occlusion:
electrocardiogram QT time and the role of collaterals. Eur Heart J.
2010;31:1197–1204.
42. Suzuki M, Nishizaki M, Arita M, Ashikaga T, Yamawake N, Kakuta T,
Numano F, Hiraoka M. Increased QT dispersion in patients with vaso-
spastic angina. Circulation. 1998;98:435–440.
43. Vogel R, Indermühle A, Seiler C. Determination of the absolute per-
fusion threshold preventing myocardial ischaemia in humans. Heart.
2007;93:115–116.
44. Hirai T, Fujita M, Nakajima H, Asanoi H, Yamanishi K, Ohno A,
Sasayama S. Importance of collateral circulation for prevention of
left ventricular aneurysm formation in acute myocardial infarction.
Circulation. 1989;79:791–796.
45. Habib GB, Heibig J, Forman SA, Brown BG, Roberts R, Terrin ML,
Bolli R. Influence of coronary collateral vessels on myocardial infarct
size in humans. Results of phase I thrombolysis in myocardial infarction
(TIMI) trial. The TIMI Investigators. Circulation. 1991;83:739–746.
46. Kodama K, Kusuoka H, Sakai A, Adachi T, Hasegawa S, Ueda Y,
Mishima M, Hori M, Kamada T, Inoue M, Hirayama A. Collateral chan-
nels that develop after an acute myocardial infarction prevent subsequent
left ventricular dilation. J Am Coll Cardiol. 1996;27:1133–1139.
47. Elsman P, van ‘t Hof AW, de Boer MJ, Hoorntje JC, Suryapranata H,
Dambrink JH, Zijlstra F; Zwolle Myocardial Infarction Study Group.
Role of collateral circulation in the acute phase of ST-segment-elevation
myocardial infarction treated with primary coronary intervention. Eur
Heart J. 2004;25:854–858.
48. Sorajja P, Gersh BJ, Mehran R, Lansky AJ, Krucoff MW, Webb J, Cox
DA, Brodie BR, Stone GW. Impact of collateral flow on myocardial re-
perfusion and infarct size in patients undergoing primary angioplasty for
acute myocardial infarction. Am Heart J. 2007;154:379–384.
49. Waldecker B, Waas W, Haberbosch W, Voss R, Wiecha J, Tillmanns H.
Prevalence and significance of coronary collateral circulation in patients
with acute myocardial infarct. Z Kardiol. 2002;91:243–248.
50. Helfant RH, Vokonas PS, Gorlin R. Functional importance of the human
coronary collateral circulation. N Engl J Med. 1971;284:1277–1281.
 728  Circ Cardiovasc Interv  December 2013
51. Meier P, Hemingway H, Lansky AJ, Knapp G, Pitt B, Seiler C. The im-
pact of the coronary collateral circulation on mortality: a meta-analysis.
Eur Heart J. 2012;33:614–621.
52. Antoniucci D, Valenti R, Moschi G, Migliorini A, Trapani M, Santoro
GM, Bolognese L, Cerisano G, Buonamici P, Dovellini EV. Relation be-
tween preintervention angiographic evidence of coronary collateral circu-
lation and clinical and angiographic outcomes after primary angioplasty or
stenting for acute myocardial infarction. Am J Cardiol. 2002;89:121–125.
53. Steg PG, Kerner A, Mancini GB, Reynolds HR, Carvalho AC, Fridrich
V, White HD, Forman SA, Lamas GA, Hochman JS, Buller CE; OAT
Investigators. Impact of collateral flow to the occluded infarct-related
artery on clinical outcomes in patients with recent myocardial infarc-
tion: a report from the randomized occluded artery trial. Circulation.
2010;121:2724–2730.
54. Williams DO, Amsterdam EA, Miller RR, Mason DT. Functional signifi-
cance of coronary collateral vessels in patients with acute myocardial in-
farction: relation to pump performance, cardiogenic shock and survival.
Am J Cardiol. 1976;37:345–351.
55. Nestico PF, Hakki AH, Meissner MD, Bemis CE, Kimbiris D, Mintz
GS, Segal BL, Iskandrian AS. Effect of collateral vessels on prognosis
in patients with one vessel coronary artery disease. J Am Coll Cardiol.
1985;6:1257–1263.
56. Hansen JF. Coronary collateral circulation: clinical significance and in-
fluence on survival in patients with coronary artery occlusion. Am Heart
J. 1989;117:290–295.
57. Pérez-Castellano N, García EJ, Abeytua M, Soriano J, Serrano JA,
Elízaga J, Botas J, López-Sendón JL, Delcán JL. Influence of collateral
Downloaded from http://ahajournals.org by on April 23, 2021
circulation on in-hospital death from anterior acute myocardial infarc-
tion. J Am Coll Cardiol. 1998;31:512–518.
58. Nicolau JC, Nogueira PR, Pinto MA, Serrano CV Jr, Garzon SA. Early
infarct artery collateral flow does not improve long-term survival follow-
ing thrombolytic therapy for acute myocardial infarction. Am J Cardiol.
1999;83:21–26.
59. Monteiro P, Antunes A, Gonçalves LM, Providência LA. Long-term
clinical impact of coronary-collateral vessels after acute myocardial in-
farction. Rev Port Cardiol. 2003;22:1051–1061.
60. Meier P, Gloekler S, Zbinden R, Beckh S, de Marchi SF, Zbinden S,
Wustmann K, Billinger M, Vogel R, Cook S, Wenaweser P, Togni M,
Windecker S, Meier B, Seiler C. Beneficial effect of recruitable collat-
erals: a 10-year follow-up study in patients with stable coronary artery
disease undergoing quantitative collateral measurements. Circulation.
2007;116:975–983.
61. Regieli JJ, Jukema JW, Nathoe HM, Zwinderman AH, Ng S, Grobbee
DE, van der Graaf Y, Doevendans PA. Coronary collaterals improve
prognosis in patients with ischemic heart disease. Int J Cardiol.
2009;132:257–262.
62. Desch S, Eitel I, Schmitt J, Sareban M, Fuernau G, Schuler G, Thiele H.
Effect of coronary collaterals on microvascular obstruction as assessed
by magnetic resonance imaging in patients with acute ST-elevation
myocardial infarction treated by primary coronary intervention. Am J
Cardiol. 2009;104:1204–1209.
Key Words: cardiac catheterization ◼ collateral circulation ◼ coronary
circulation ◼ electrocardiography ◼ microcirculation ◼ perfusion