Article Text

Download PDFPDF

Coronary flow: clinical considerations
  1. D V Cokkinos1,
  2. A Manginas2,
  3. V Voudris2
  1. 1University of Athens, Onassis Cardiac Surgery Center, Athens, Greece
  2. 2Onassis Cardiac Surgery Center
  1. Correspondence to:
    Professor Dennis V Cokkinos, Onassis Cardiac Surgery Center, 1st Cardiology Department, 356 Sygrou Avenue, 174 74 Athens, Greece;

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

In the measurement of coronary blood flow to determine the success of percutaneous coronary intervention, invasive techniques, coupled with plaque characterisation and other intracoronary imaging modalities, may prove invaluable

The measurement of coronary blood flow (CBF) has transcended the realm of observational curiosity through the emergence of two developments:

  • Firstly, the application of easier methods to measure it, which have replaced the older cumbersome inert gas and thermodilution techniques. Two notable examples are the non-invasive positron emission tomography (PET) calculations and the intracoronary Doppler measurements of flow velocity. Formerly, Doppler catheters were used, which could not measure flow distally to a stenosis. This problem was overcome by the use of the Flowire, which with a diameter of 0.014 inches can be placed across the stenosis during the course of invasive procedures. Because of its small diameter, this wire does not cause significant flow disturbances, as was the case with the previously used Doppler catheters. The Doppler wire actually measures flow velocity. For the velocity values to be equivalent to CBF, the cross sectional area of the vessel must be measured.

  • Secondly, the realisation that diminution of flow distally to a significant stenosis is reliably correlated to manifestations of ischaemia.

Concomitantly with the measurement of CBF, the importance of the coronary flow reserve (CFR) was appreciated. This term signifies the difference between CBF under maximal hyperaemia, produced by various interventions, and basal flow. Basal CBF amounts to approximately 1 ml/g of myocardial tissue/min; at maximal hyperaemia it can increase by 3–4 times, according to the technique employed.

Many factors influence the CFR values. Because of the use of CFR in the assessment of coronary artery stenosis, the degree of stenosis has attained major importance. However, the actual coronary resistance is determined to a far greater degree by microvascular resistance. Since the assessment of microvascular resistance, as will be discussed later, is difficult, the concept of the relative coronary flow reserve (RCFR) has emerged, in which the CFR of the involved artery is compared to an adjacent normal artery. Thus each patient is his own control. A not sufficiently appreciated element is that a high basal CBF amounts to a lower CFR. This old concept is “discovered” anew whenever a new technique measuring coronary flow emerges.

Gould and his team showed in 1974 in the dog that while basal coronary flow remains normal until a stenosis ≥ 90% is produced, hyperaemia is abolished above a coronary artery diameter stenosis ≥ 60–70%.1 This finding, later documented with PET measurements, clarified the mechanisms of effort angina; during physical exercise the pressure–rate product, which corresponds closely to myocardial oxygen consumption, can increase up to three times, while CBF does not increase.

When CBF is assessed to characterise the impact of a coronary stenotic lesion, a single measurement does not suffice. According to Kern, a significant lesion is associated with one or more of the following abnormalities: a post-stenotic absolute CFR < 2.0, RCFR < 0.9, proximal to distal flow velocity ratio < 1.7, diastolic to systolic velocity ratio < 1.8, and when the fractional flow reserve (FFR) is used, a value < 0.75.2

FFR is another index used to assess the functional significance of coronary stenoses. It is calculated as the ratio of the pressure distal to the stenosis during maximal hyperaemia to the aortic pressure, measured by a pressure sensor mounted on a 0.014 inch guide wire. Values < 0.75 are considered haemodynamically significant.

Correlations of CFR with ischaemic manifestations were prompt to emerge. Wilson and colleagues3 showed that at exercise electrocardiography, the degree of ST depression corresponded significantly with the CFR but not the severity of epicardial artery stenosis. Bartunek and colleagues4 using dobutamine echocardiography also showed that wall motion abnormalities emerged with a diameter stenosis ∼60% and an FFR ∼0.67. Our group was the first to show that the degree of Tl-201 reversibility correlated very strongly with the difference of CFR between a normal artery and an adjacent artery with a significant stenosis.5

We were also able to offer some insight into why adenosine (and its equivalent dipyridamole) are preferred over dobutamine for the pharmacologic stress of Tl-201 scintigraphy. We showed that adenosine produces flow inhomogeneity to a greater degree than maximal dose dobutamine alone.6

FFR values < 0.75 have also been correlated with ischaemia, both at exercise testing and myocardial perfusion imaging.7

A question which arises when CBF is measured is whether the large epicardial arteries or the microvasculature are at fault. The main disease entities associated with microvascular disease are a previous myocardial infarction, left ventricular hypertrophy, diabetes mellitus, the syndrome X and others. The pressure wire cannot address this question. A low CFR in the presence of normal FFR is a practical index of microvascular dysfunction. Very recently Pijls and colleagues8 combined FFR and CFR by coronary thermodilution to facilitate assessment of microvascular disease.


The DEBATE trial evaluated the effect of balloon angioplasty on distal CFR in relation to the clinical outcome during a six month follow up period.9 Postprocedure CFR was found to have a modest prognostic value in predicting the incidence of symptoms and/or ischaemia at four weeks. When anatomic and physiologic data were combined, a distal CFR after balloon angioplasty > 2.5 with a residual diameter stenosis ≤ 35% identified lesions with a low incidence of recurrence of symptoms at one month and six months, a low need for reintervention, and a lower restenosis rate compared with patients who did not meet these criteria. In the DEBATE II study, a low CFR postprocedure was an independent predictor of major adverse cardiac events at 30 days and at one year.10

The DESTINI study11 indicated that, compared with elective stenting (n = 370 patients), provisional stenting (n = 218 lesions) based on CFR and angiographic criteria can avoid the costs and complications of stenting in about 50% of patients with no compromise in clinical or angiographic outcomes.

Bech and colleague12 analysed the prognostic significance of FFR in patients undergoing balloon angioplasty. The two year event-free survival for patients with postprocedural diameter stenosis ≤ 35% and FFR ≥ 0.90 was significantly better than that for patients with suboptimal values of either of these variables. FFR immediately after stenting was the most significant independent variable related to major adverse cardiac events, including the need for repeat target vessel revascularisation at six months.

An inverse situation was also considered by Kern13 in the evaluation of CBF measurements and whether it could be used to defer a percutaneous coronary intervention (PCI). He found a low rate of clinical events (< 10%) over a two year follow up period in intermediate stenoses with a normal CFR. Very recently Chamuleau and colleagues14 found that a CFR< 2.0 was a significant predictor of events—stronger than Tl-201 SPECT. Actually they found few events (1/21) at one year follow up in patients with a positive SPECT and a normal CFR. We do not know how many clinicians would decide to measure CFR rather than proceed to PCI in a patient with a strongly positive SPECT who is found to have a 60% lesion at angiography.


For the last nine years, studies following acute myocardial reperfusion either by thrombolysis or percutaneous interventions after an acute occlusion are associated with appreciable coronary flow abnormalities. CFR gives a measure of the anticipated myocardial recovery. Gibson and colleagues15 showed that improved epicardial flow at 90 minutes after thrombolysis was associated with improved survival. Stone and associates16 reported the same results with primary or rescue angioplasty.

Gibson and colleagues17 were the first to advance another entity in CBF, the tissue level perfusion assessed by the myocardial perfusion grade, which is expressed by the “myocardial blush”.

Myocardial hibernation refers to a state of chronic ischaemic ventricular dysfunction characterised by improvement after coronary revascularisation. By definition, viability of the cardiomyocytes is preserved as a result of metabolic downregulation, an important adaptive mechanism against myocardial necrosis. A controversy exists as regards the reduction of resting myocardial flow in hibernating segments and its clinical impact, reflected also by the use of various terms such as “chronic stunning”, and “short term” and “long term hibernation”. In a meta-analysis published in 1999, Schwartz and colleagues18 did not find convincing evidence supporting the widely accepted notion of chronic basal flow reduction in hibernating myocardium.

It appears that the differences in various studies are due to observations of a progressive pathophysiological phenomenon at various points along its natural course. During the initial stages the resting blood flow is not necessarily reduced, the post-revascularisation contractile recovery is complete, but the CFR in the hibernating segments is clearly diminished (early hibernation).19,20 As coronary stenoses progress, the episodes of ischaemia and stunning become more frequent (repetitive stunning), gradually resulting in functional deterioration and possibly eventual depression of the resting myocardial flow, followed by cellular alterations and incomplete functional post-revascularisation recovery (late hibernation).21


Flowire and the pressure wire techniques are costly and invasive. PET is non-invasive, but costly and not widely available. The Tl-201 washout rates have not been widely correlated nor widely used to assess CBF, and contrast echocardiography is only now coming of age for this purpose.

Transoesophageal echocardiography has been used for more than 10 years to measure CFR with consistently good correlations. However, the technique is cumbersome. Newer echocardiographic techniques have enabled the measurement of CBF by transthoracic echocardiography. Correlations with invasive techniques have proved very good.22 Results may further improve with contrast echocardiography.

A corresponding technique has been the Doppler measurement of flow in the left internal mammary artery (LIMA) grafted to the left anterior descending (LAD) coronary artery, from the left supraclavicular fossa. We have found that the CFR values with this technique correspond very closely (r = 0.88) with those obtained by use of the Flowire, and that a normal LIMA/LAD CFR has an excellent prognostic value.23

An invasive but inexpensive method was devised by Gibson and colleagues,24 the TIMI frame count—that is, the time required for contrast material to reach certain landmarks of coronary anatomy.

Our group was the first to show that CFR measured with the TIMI frame count after intravenous adenosine administration has an excellent correlation with that measured with the Flowire.25 We have also used the baseline TIMI frame count post-angioplasty administration to predict restenosis after PTCA. A ratio of a TIMI frame count/minimal lumen diameter was strongly predictive of restenosis.26 This was very recently verified by Gibson and colleagues27 for PTCA or directional atherectomy in acute coronary syndromes. If this index is found to be true for stenting also, this technique can provide a simple means to guide interventional efforts at the catheterisation table.


Conjecture is a fruitless pastime. However, we can venture some predictions. Non-invasive techniques will probably predominate in the coming years. Flow of the epicardial arteries by Doppler, microvascular function by contrast injection, and tissue viability by Doppler tissue imaging will be combined in one echocardiographic measurement. The same can probably be achieved by magnetic resonance imaging, which is costly and cumbersome however.

These two techniques will be increasingly and interchangeably used to assess success of PCI in acute—especially myocardial infarction—and chronic situations to examine tissue viability. Non-invasive CFR measurement reflecting microvascular function will direct and monitor medical treatment in unfavourable metabolic situations, such as scar tissue, the metabolic syndrome, diabetes mellitus, dyslipidaemia, hypertension, and dilated cardiomyopathy.

However, we believe that it is unwise to sing a premature requiem for the invasive techniques. If they are coupled with plaque thermography or other intracoronary imaging modalities such as optical coherence tomography or refined intravascular ultrasound systems, they may prove invaluable for early detection of the unstable plaque, an obsession of the current cardiologist comparable to the quest of the medieval knights for the holy grail.

In the measurement of coronary blood flow to determine the success of percutaneous coronary intervention, invasive techniques, coupled with plaque characterisation and other intracoronary imaging modalities, may prove invaluable


View Abstract