Article Text

other Versions

The coronary collateral circulation—clinical relevances and therapeutic options
  1. Pascal Meier1,2,
  2. Christian Seiler3
  1. 1Department of Cardiology, The Heart Hospital, University College London, London, UK
  2. 2Department of Cardiology, Yale Medical School, New Haven, Connecticut, USA
  3. 3Department of Cardiology, University Hospital Bern, Bern, Switzerland
  1. Correspondence to Professor Christian Seiler, Department of Cardiology, University Hospital Bern, Freiburgstrasse, Bern 3000, Switzerland; christian.seiler{at}

Statistics from

The coronary arteries were once thought of as end-arteries. Certainly, they often behave like functional end-arteries, as illustrated by the ischaemia induced by single vessel coronary artery disease. However, we now know that there are interconnecting branches between the main arteries although their clinical relevance has been disputed, since the anastomoses are often incapable of restoring flow to normal levels (figure 1).

Figure 1

Left side: Heart with well-developed coronary collateral circulation and, therefore, much smaller area at risk compared with the heart on the right side with poorly developed collaterals. (Illustration by Anne Wadmore, Medical Illustrations Ltd, UK).

Hitherto, 12 studies have investigated the effect of collaterals on survival, the first in 1971,1 but only three have demonstrated a clear benefit leaving unresolved the dispute about their functional relevance.2 The inconsistency is partially explained by the method of assessment used in most of the studies whereby collaterals were ‘quantified’ visually during coronary angiography. This represents a rather crude approach compared with the more accurate intracoronary flow or pressure-based methods (collateral flow index) using Doppler or pressure sensors applied to the tip of a guidewire.3 However, in certain situations, the relevance of the collateral circulation is obvious, the preservation of ventricular function in some patients with chronic total coronary occlusions providing a good example of this. There are even extreme examples of patients with left main artery occlusion, or three-vessel occlusion, with only mild symptoms;2 but this anecdotal evidence has now been supplemented by a pooled analysis of the above mentioned 12 studies, including 6529 patients, that has shown that well-developed collaterals are associated with a reduced mortality of about 35%.3

The degree of collateralisation varies considerably among patients. Several clinical factors are known to play a role, particularly the degree of stenosis which consistently emerges as the most important in angiographic studies.4 In addition, some studies have found an association with proximal lesion location, longer duration of angina, longer duration of a chronic total occlusion and with bradycardia.5 Even though degree of stenosis is a strong predictor, collaterals are also present in patients without coronary artery disease. Interestingly, the collateral circulation is not affected after heart transplantation.6

Multiple strategies to enhance collateral function have been tested, such as administration of growth factors, progenitor cells and physical measures, such as exercise and external counterpulsation. While several of them have worked in animal models, most have failed in first-in-man studies with a few promising exceptions (table 1).

Table 1

Factors that can improve the collateral circulation

Current understanding is that collateral growth (called arteriogenesis) happens via a remodelling process of pre-existing small collaterals (collateral remodelling). It differs from angiogenesis, the growth of new capillary vessels, which is induced by ischaemia.13 Collateral growth, on the other hand, is induced by fluid shear stress in preformed collateral vessels caused by a pressure gradient between the area proximal to a coronary stenosis and the low-pressure poststenotic area. The shear stress on endothelial cells stimulates the production of nitric oxide and monocyte chemoattractant protein 1, leading to an attraction of monocytes which play a key role in orchestrating collateral remodelling, including attraction of endothelial progenitor cells.14

The important roles of shear stress and of monocytes have been used as targets for the therapeutic induction of collaterals. Granulocyte macrophage colony stimulating factor and Granulocyte colony stimulating factor are growth factors that increase monocyte numbers, and they have both been shown to improve collateral function.11 Another therapeutic option is to increase sheer stress by external counterpulsation8 or physical exercise,15 both of which have been shown to improve collateral function. Another promising approach for increasing shear stress is treatment with ivabradine. Bradycardia is known to be associated with better collateralisation (table 1), probably because the lower heart rate increases endothelial shear stress. Experimental studies have indicated a benefit of ivabradine on collateral growth.16 A clinical study to test this concept in humans is currently in progress ( identifier NCT01039389).

Interestingly, several of the above mentioned studies have shown a mortality benefit for collaterals in patients undergoing percutaneous coronary intervention (PCI),3 in line with data on coronary bypass grafting (CABG) and its mortality benefit over medical therapy or PCI in patients with coronary artery disease.17 ,18 Analagous to bypass surgery, collaterals can be regarded as a natural bypass system which protects the myocardium against ischaemia. While PCI only treats the ‘culprit lesion’ (target lesion), bypass grafts (and similarly the collateral network), protect almost the entire target vessel and its perfusion territory.

In conclusion, coronary collaterals seem to provide a significant mortality benefit, even in patients undergoing percutaneous revascularisation. This is in line with recent data demonstrating a survival benefit of CABG over PCI. Therapeutic induction of collateral growth is a potentially attractive complementary therapy for PCI to overcome this shortcoming.

View Abstract


  • Contributors PM drafted the text, CS revised the text critically and both authors approved the final version of this text. Both authors contributed substantially to this text.

  • Funding This project was supported by the Swiss National Science Foundation (grant number 3200BO-112341 to CS.

  • Competing interests None.

  • Provenance and peer review Commissioned; internally peer reviewed.

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.