Article Text

Download PDFPDF

Should radial artery access be the “gold standard” for PCI?
  1. Martial Hamon1,
  2. James Nolan2
  1. 1
    University Hospital of Caen, Caen, France
  2. 2
    University Hospital of North Staffordshire, Stoke-on-Trent, UK
  1. Professor M Hamon, Service des Maladies du Coeur et des Vaisseaux, UF Soins Intensifs Cardiologiques, Centre Hospitalier Universitaire de Caen, Avenue Côte de Nacre 14033 Caen, Normandy, France, Cedex; hamon-m{at}

Statistics from

Consider this situation. A 65-year-old patient admitted to hospital with a troponin-positive acute coronary syndrome (ACS) is initially managed with intensive antithrombotic therapy. Angiography is performed via the femoral artery, demonstrating a tight thrombus-containing lesion in the left anterior descending artery that involves the first diagonal branch. Percutaneous coronary intervention (PCI) is performed, with a stent implanted into the left anterior descending artery, using a kissing balloon procedure to achieve an excellent final result. After this successful and uneventful procedure, the patient is transferred back to the cardiology ward. On arrival at the ward it is noted that the patient has a low blood pressure and complains of groin discomfort. Inspection shows an extensive haematoma at the femoral puncture site. A large and expanding haematoma is demonstrated by ultrasound examination with an associated fall in haemoglobin to 7.2 g/dl. An immediate transfusion of three units of blood is administered and vascular surgical repair is required. Is this a rare, inconvenient, benign and unavoidable component of contemporary ACS management? The MORTAL study, published recently in Heart,1 complements an extensive and rapidly evolving literature that can help to answer these questions.

The prevalence of PCI-related bleeding depends to some extent on the study population and the bleeding definition employed.2 In patients with ACS, the rate of major bleeding is between 3 and 5%.2 3 Femoral access site complications are responsible for most of the bleeding that occurs in invasively managed patients with ACS. The complications of femoral artery access, including hematomas, arteriovenous fistulae, arterial pseudoaneurysms and retroperitoneal haemorrhage, are influenced by anatomical variations, obesity and puncture technique. Glycoprotein IIb/IIIa inhibitors, the use of which are recommended with a class I level of evidence in current ACS guidelines,4 5 further increase the risks of perivascular and systemic bleeding. In patients where femoral artery access is used, the sheath is often removed several hours after the procedure to allow coagulation recovery after discontinuation of antithrombotic therapy, a situation that may increase bleeding risk. Alternatively, femoral sheaths can be removed immediately in conjunction with the use of vascular closure devices. Unfortunately, vascular closure devices have not been found to reduce the rate of haemorrhagic or vascular complications in a meta-analysis of randomised trials.6 7

The use of improved antithrombotic strategies has reduced the incidence of cardiac ischaemic events in patients with ACS who are treated either conservatively or with PCI. The downside of intensive antithrombotic therapy is a significant increase in the risk of bleeding complications. Major bleeding complications are now more common than cardiac ischaemic events in contemporary ACS and PCI populations. A causal relationship between major bleeding in patients with ACS and/or PCI and increased mortality and morbidity has recently been demonstrated in several publications.3 4 8 9 The early recognition of bleeding risk in patients with ACS, especially those scheduled to have PCI, is of critical importance in optimising patient outcomes (tables 1 and 2). Preprocedural assessment of bleeding risk is possible and should be taken into account when defining the risk/benefit ratio of an invasive strategy.

Table 1 Identification of risk factors for bleeding
Table 2 Identification of preventive actions to reduce bleeding risk

The mechanisms that underlie the increased mortality in patients with major bleeding are unclear, but may relate to an increased propensity of these patients to develop life-threatening ischaemic events. These deleterious effects may be due to the adverse effect of major bleeding on renal function, myocardial ischaemia or platelet activation. Alternatively, transfusion of stored blood and its associated inflammatory responses may explain this observation.8 9 In keeping with this, transfusion is the strongest predictor of length of in-hospital stay after PCI and is associated with increased mortality in patients with underlying coronary artery disease.8 9

Bleeding and anaemia also lead to discontinuation of aspirin or thienopyridines, or both, causing ischaemic complications that may also explain the increase in morbidity and mortality.10 Although the underlying mechanisms are uncertain, haemorrhagic complications and transfusions have been identified as independent predictors of adverse outcomes and significantly influence combined efficacy and safety end points, as demonstrated in the recent OASIS 5 and ACUITY trials.10 11 These data suggest that pharmacological and technological advances have now reduced ischaemic complications to the point that reduction in bleeding events should become the primary target to further improve PCI outcomes.1012

In addition to the choice of antithrombotic agents employed, vascular access site selection can also have an impact on bleeding complications. Radial artery access, compared with femoral artery access, has been associated with a reduced risk of access site bleeding and other vascular complications in meta-analysis of randomised trials.13 The MORTAL study, published recently in Heart, retrospectively examined the association between access site, transfusion, and outcomes in over 32 000 patients who underwent PCI in British Columbia from 1999 to 2005.1 The main finding was that by reducing vascular access site complications, the use of the radial access site was associated with a 50% reduction in transfusion rate and a relative reduction in 30-day and 1-year mortality of 29% and 17%, respectively (p<0.001), which corresponds to around 1% absolute risk reduction at 1 year. Therefore, the number needed to treat by the radial approach was 100 patients to save one life.

A recent meta-analysis (table 3) of bleeding in patients with ACS reported an 11% increase in the absolute risk of death associated with major bleeding (95% CI 8% to 14%), which corresponds to a number needed to harm of only 9.12 This implies that by avoiding nine major bleeds one life is saved, which is very close to the NNH of 7.4 found in the MORTAL study at 1 year. The MORTAL data are consistent with the recently reported RIVIERA study, a large prospective international registry, which also reported that use of radial access, by limiting bleeding risk and transfusion requirement, is associated with a reduction in PCI-related mortality.14 These are important messages for practising interventional cardiologists, suggesting that the radial approach should be considered as the preferred access site whenever possible. This strategy has proven benefits in improving patient comfort and cost reduction. The MORTAL and RIVIERA data suggest that there is also a beneficial effect on mortality achieved through avoidance of access site bleeding.

Table 3 Mortality in patients with or without major bleeding: risk difference and numbers needed to harm (adapted from reference)

There are some caveats related to this interpretation of the MORTAL data. The retrospective and observational design of the study mandates a cautious approach to the results. Since access site selection was not randomly assigned, there is significant opportunity for bias to be introduced. Even after multivariable adjustment to account for baseline differences, the extent to which unmeasured confounders can affect the propensity for operators to choose the radial over the femoral access site cannot be determined in this type of observational study. Continuing randomised trials (a substudy of the CURRENT trial and the RIVAL trial) will ultimately confirm or refute these findings. Even with these important cautions and caveats, the MORTAL results are entirely consistent with the extensive body of existing data indicating that radial access limits bleeding risk, and this would be expected to improve outcome. The benefits of radial access may be further enhanced by employing drug strategies that maintain antithrombotic efficacy but limit bleeding risk11 15 (tables 1 and 2).

The transradial technique requires a specific set of skills, and is associated with a significant learning curve. Radial artery spasm, arterial puncture failure, vascular anomalies and failure to reach the ascending aorta can be obstacles that impede widespread uptake of this approach. With appropriate training, however, comparable success rates with the radial and femoral approaches may be achieved even in complex cases,1618 with associated reduced rates of haemorrhagic and vascular complications,13 19 20 thus justifying the more widespread use of this technique as supported by the MORTAL study.1

The answers to the questions posed in our first paragraph are that important bleeding after PCI is common, particularly when femoral access is employed in the setting of intensive antithrombotic therapy. This bleeding is not inconvenient or benign as it is often associated with major adverse events. The data available from randomised trials suggest that interventional cardiologists can avoid this risk by switching their choice of access site and using the radial artery whenever possible. The past 50 years have seen cardiologists switch from a surgical approach to the brachial artery towards a technically simpler percutaneous femoral approach. Interventional cardiologists now have an opportunity to evaluate the MORTAL data in the context of current knowledge of the adverse effect of bleeding on PCI outcomes. The evidence base that is available supports a switch to radial access for most PCI procedures with the aim of improving outcome by reduction in access site bleeding.


View Abstract


  • Competing interests: None.

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.

Linked Articles