Article Text

Download PDFPDF

Original article
Routine use of the transradial approach in primary percutaneous coronary intervention: procedural aspects and outcomes in 2209 patients treated in a single high-volume centre
  1. Maarten A Vink1,
  2. Giovanni Amoroso1,
  3. Maurits T Dirksen1,
  4. Rene J van der Schaaf1,
  5. Mark S Patterson1,
  6. Jan G P Tijssen2,
  7. Ferdinand Kiemeneij1,
  8. Ton Slagboom1
  1. 1Department of Cardiology, Onze Lieve Vrouwe Gasthuis, Amsterdam, The Netherlands
  2. 2Academic Medical Centre, Amsterdam, The Netherlands
  1. Correspondence to Dr Maarten A Vink, Department of Cardiology, Onze Lieve Vrouwe Gasthuis, Oosterpark 9, 1091 AC Amsterdam, The Netherlands; m.a.vink{at}


Objective To examine the feasibility of a routine transradial approach (TRA) in primary percutaneous coronary intervention (PPCI) for acute ST-segment elevation myocardial infarction (STEMI).

Design A single-centre observational study with prospective data collection.

Setting A high-volume interventional centre in Amsterdam, The Netherlands.

Patients Procedural data were analysed for 2209 consecutive patients presenting with STEMI without cardiogenic shock, between January 2001 and December 2008.

Interventions PPCI routinely performed by the TRA.

Main Outcome Measures The primary outcomes of interest were the need for crossover to another vascular access site, the achievement of procedural success and their trends over time. Secondary outcome measures were trends in total procedural duration, fluoroscopy times and use of equipment.

Results In a total of 2209 procedures the radial artery was the primary access site, comprising 96.1% of all procedures performed during the study period. In 84 cases (3.8%) access site crossover was needed. Crossover rates decreased from 5.9% in 2001–2 to 1.5% in 2007–8 (p=0.001). The procedural success rate was 94.1%, which remained stable over the years. Despite an increased complexity of PPCI (more non-left anterior descending infarct-related arteries, thrombus aspiration and multivessel PPCI), total procedural duration decreased from 38 min (IQR 28–50) in 2001–2 to 24 min (18–33) in 2007–8, p<0.001 for trend.

Conclusions Systematic use of the TRA in PPCI yields low access site crossover, high procedural success rates and excellent procedural performances. It can therefore represent the primary access site in the vast majority of STEMI patients.

  • Acute coronary syndrome
  • acute ischaemic syndromes
  • acute myocardial infarction
  • angiography
  • coronary intervention (PCI)
  • coronary stenting
  • electrocardiography
  • interventional cardiology
  • intravascular ultrasound
  • primary PCI
  • primary percutaneous coronary intervention
  • reperfusion injury
  • transradial access
View Full Text

Statistics from

Primary percutaneous coronary intervention (PPCI) is the treatment of choice in patients presenting with acute ST-segment elevation myocardial infarction (STEMI). PPCI gives excellent results in terms of successful reperfusion, but also bears the risk of bleeding complications, which occur at the site of arterial access in a considerable proportion of cases.1 2 Access site-related complications have a negative impact not only on morbidity but also on mortality after PPCI.2 3 The transradial approach (TRA) may help to reduce access site complications after PPCI, as a few studies have found a reduction in vascular complications while achieving similar procedural results using TRA–PPCI, compared with the transfemoral approach.4–7 Due to the limited size of these studies and the inclusion of only highly selected patients, it was debated whether their results would be applicable to the broad population of patients undergoing PPCI. A recent study comparing TRA with the transfemoral approach in acute coronary syndromes found similar event rates with both treatment strategies, and it was suggested that TRA may be particularly beneficial in STEMI patients and when performed in high-volume centres.8 In that study, however, only a small proportion of patients underwent PPCI, and operators and centres did not pursue a systematic TRA strategy. Therefore concerns remain whether routine TRA–PPCI would be a limitation with regard to the choice of devices, procedural times and radiation exposure and procedural success rates.

After its first introduction in the early 1990s,9 TRA has become the standard of care for percutaneous coronary intervention in our institution and,10–13 for over a decade, the radial artery has represented the primary access site also in PPCI. The purpose of this study was to evaluate rates of achieving successful arterial access and procedural performances of the routine use of TRA–PPCI and their trends over a time span of 8 years.


Source population and procedures

All patients who underwent PPCI for STEMI at the Onze Lieve Vrouwe Gasthuis between January 2001 and December 2008 were included in the present analysis. In general, patients with symptoms of an acute myocardial infarction lasting less than 12 h and ST-segment elevation greater than 0.1 mV in at least two contiguous leads on the electrocardiogram had an indication for PPCI. Patients undergoing rescue percutaneous coronary intervention after failed thrombolysis were not included in this analysis.

Before the procedure, all patients received a loading dose of aspirin and clopidogrel according to the latest standards. In addition, patients received a bolus of heparin of 10 000 IU, or 5000 IU in the case of the administration of glycoprotein IIb/IIIa inhibitors. PPCI was performed according to the existing clinical practice guidelines, by interventionalists with at least 1 years' experience in TRA but also by fellows in training under supervision. The right radial artery was the preferred site for vascular access. To prevent arterial spasm, a spasmolytic cocktail composed of 200 μg nitroglycerin and 5 mg verapamil was given at the discretion of the operator. Procedures were performed with 6 F guiding catheters with curves preferably fitting both the left and the right coronary ostia. The arterial sheath was removed immediately after the procedure, and haemostasis maintained for no longer than 4 h in order to avoid radial occlusion. Thrombus aspiration catheters (mechanical or manual) were used from the end of 2001 and drug-eluting stents from the beginning of 2003.

Baseline clinical, angiographic and procedural variables were collected prospectively by our catheterisation laboratory staff and the attending interventional cardiologist, and were subsequently recorded in a dedicated database as part of routine patient care. We identified a total of 2484 procedures, of which 184 (7.4%) were excluded from further analysis because patients were in cardiogenic shock at presentation.

Data analysis

The primary outcomes of interest of the present analysis were the need for crossover to another puncture site and the achievement of acute procedural success in TRA–PPCI, as well as their trends over time. Acute procedural success was defined as a thrombolysis in myocardial infarction (TIMI) flow grade 3 or an improvement of the TIMI flow of 2 grades and 30% or less stenosis in the culprit lesion at the end of the procedure. In addition, we analysed trends in procedural times and the use of equipment during the study period. Duration of the PPCI procedure was defined as the time from the first attempt to puncture the radial artery to sheath removal. Baseline clinical characteristics and rates of crossover and acute procedural success are expressed as frequencies (percentage) or mean±SD. Temporal trends in categorical variables were compared using the χ2 test, while trends in procedural times and contrast use were assessed using the Kruskal–Wallis test.

Furthermore, we determined which clinical and procedural variables were associated with the need for crossover, or with procedural failure, by the use of binary logistic regression analysis in both univariable and multivariable models. Variables were left out in case of missing values in more than 10% of patients. The results were expressed as OR with 95% CI. All tests were two-tailed, and a value of p<0.05 was considered statistically significant. Statistical analysis was performed with SPSS software (SPSS V.18.0 for Windows).


Study cohort

Between January 2001 and December 2008, 2300 PPCI for STEMI were performed on 2232 patients. The femoral artery was chosen as the primary access site in 91/2300 cases (3.9%). The most frequent reasons to refrain from TRA were previous radial access within hours before the procedure (ie, in the case of acute stent thrombosis) in 18/91 cases (19.7%), and anatomical impediments (ie, broken arm or Cimino shunt) in 17/91 cases (18.6%), see table 1. The femoral artery was chosen as the primary access site primarily in 2001–2 (36/392 procedures, 9.2%), after which in 2007–8 the femoral artery was chosen in only eight of 616 cases (1.3%), p<0.001 for trend.

Table 1

Incidence of transfemoral PPCI and motives to refrain from TRA–PPCI

Crossover in TRA-PPCI

A total of 2209 TRA-PPCI was attempted during the study period, comprising 96.1% of all PPCI procedures performed.

Patients' clinical characteristics are shown in table 2. The minority of patients was women (25.0%) and patients had a mean age of 61613 years. In a total of 44 procedures (2.1%), patients had a history of coronary artery bypass grafting. From 2001–2 to 2007–8 a slight increase in patients with a history of hypertension was observed (25.6% in 2001–2, 34.4% in 2007–8, p=0.02). Furthermore, we observed a reduction in anterior wall infarctions (57.3% to 39.1%, p<0.001), otherwise there were no significant changes in baseline demographic and clinical variables.

Table 2

Demographic and clinical characteristics of the study population

In 84/2209 (3.8%) of the attempted TRA-PPCI, crossover to another access site (8/84 brachial and 76/84 femoral artery) was required. The most frequent reason for crossover was the inability of the wire and/or guiding catheter to reach the coronary ostium due to peripheral occlusion or tortuosity of the brachiocephalic trunk in 30/84 cases (35.7%) and inadequate catheter support in 14/84 cases (16.7%), see table 3.

Table 3

Causes of failure to proceed with TRA–PPCI in 84 procedures (3.8% of total)

From the years 2001–2 to 2007–8, the need for crossover decreased from 5.9% to 1.5% (p<0.001 for trend), see table 4.

Table 4

Angiographic characteristics and procedural performances of the study population

After multivariable adjustment, three independent predictors of the need for crossover were identified: female gender (OR 2.04, 95% CI 1.26 to 3.32, p=0.004), age above 75 years (OR 1.72, 95% CI 1.00 to 2.94, p=0.049) and a history of coronary artery bypass grafting (OR 4.38, 95% CI 1.57 to 12.23, p=0.005).

Angiographic characteristics and procedural performances

Angiographic characteristics and procedural performances are summarised in table 4. A spasmolytic cocktail was used in 72.8% of cases and became standard treatment since 2003–4. There was a significant shift in the infarct-related artery; while the occurrence of left anterior descending artery (LAD) occlusions remained substantially stable in absolute values (203 vs 233 in 2001–2 and 2007–8, respectively), right coronary artery and left circumflex as infarct-related artery (IRA) increased from 121 to 270 and from 26 to 99 between 2001–2 and 2007–8 (p<0.001 for both). As a result, LAD (57.0%) and non-LAD vessels (61.7%) formed the majority of infarct-related arteries in 2001–2 and in 2007–8, respectively. Multivessel PPCI increased from 2.5% to 4.6% (p=0.03).

The use of glycoprotein IIb/IIIa inhibitors decreased from 89.3% in 2001–2 to 16.9% in 2007–8 (p<0.001), while thrombus aspiration increased from 27.0% to 65.3% (p<0.001), becoming predominant as of 2003–4. The mean number of guiding catheters used was 1.23±0.61, and a strategy of using only one catheter for both coronary arteries ensued in 83.0% of all procedures. In 93.8% of cases a stent was inserted, at an average of 1.28±0.72 stents per procedure, with a mean use of 20.5% of drug-eluting stents (since 2003–4). Both stenting and the number of catheters used did not change significantly over the years, see table 4.

Acute procedural success was achieved in 2078/2209 (94.1%) of all TRA–PPCI, with no significant changes from 2001–2 to 2007–8. The procedural success rate was significantly higher in the case of TRA success versus crossover (93.6% vs 88.1%, respectively, p=0.044) but, after multivariable adjustment, crossover was not a predictor of procedural failure (OR 1.47, 95% CI 0.66 to 3.25, p=0.35).

The median duration of the procedures was 30 min (IQR 21–42), while the median time of fluoroscopy was 8.2 min (IQR 5.4–12.7) and the average contrast use was 200 ml (IQR 150–250). From 2001–2 to 2007–8 total procedural time, fluoroscopy time and contrast use declined, respectively, from 38 (28–50) to 24 min (18–33), from 9.7 (6.5–14.6) to 6.4 min (4.3–10.2) and from 250 (200–350) to 150 ml (110–200), p<0.001 for all, see table 4.

Procedural and fluoroscopy times were significantly lower in the case of TRA success versus crossover, at 29 min (IQR 21–41) versus 46 min (IQR 37–62) and 8.1 min (5.4–12.5) versus 12.6 min (8.2–18.0), respectively, p<0.001 for both.


Provided TRA is systematically pursued, the present study demonstrates that when TRA is used for PPCI, both very high success rates in vascular access and excellent procedural outcomes can be achieved. In fact, even though TRA was attempted in nearly all PPCI performed in our 8-year experience (2209/2300 procedures, 96.1% of total PPCI), the crossover rate was only 3.8%. In spite of the routine use of TRA and the growing complexity of the procedures performed over the years, acute procedural success rates remained very high at an average of 94.1%, while the parameters of procedural performance (procedural and fluoroscopy times) improved continuously.

TRA–PPCI and the need for crossover

In the past years, a few small-sized randomised trials comparing TRA with transfemoral PPCI reported crossover rates between 0% and 12.3%.4–7 Three recent retrospective studies reported a TRA failure of 7.7%, 6.5% and 4.7%, respectively, in the setting of a heterogeneous population including STEMI.14–16 Although all those studies strongly favoured TRA over transfemoral PPCI in terms of access site complications, they did not generate a significant change in clinical practice,17–19 mostly because they could not detect significant differences in hard endpoints (mortality or major adverse cardiac events), but also because they had substantial selection bias with regard to patient selection as well as the experience and dedication of the operators. For instance, in the reports of Hetherington et al14 and Dehghani et al,16 TRA accounted for only 54% and 38% of the procedures performed, respectively. Unfortunately, two very large retrospective studies in an all-comer population including STEMI did not report their rates of crossover nor the rates of procedural success.20 21 Finally, in the recently published Radial versus femoral access for coronary angiography and intervention in patients with acute coronary syndromes (RIVAL) study,8 crossover for TRA failure was still needed in 7.8% of the whole population with acute coronary syndrome and in 5.3% of STEMI patients. Of note, the crossover rate was only 4.4% in the highest tertile by TRA volume.

Our study confirms that routine TRA can be highly successful in the setting of PPCI in a broader population. In our experience, not only was TRA successful in the majority of procedures (96.2%), but access site crossover also underwent a substantial decrease from 5.9% to 1.5% during the study period, while the percentage of procedures in which TRA was not the primary approach dropped from 9.2% to 1.3%. In a large study involving retrograde radial arteriography, it was recently shown that an anomalous radial artery anatomy is relatively common and thereby may cause access failure.22 In the present study, and in agreement with other studies, female gender, age greater than 75 years and saphenous vein graft (SVG) interventions were independent predictors of crossover.16 However, as female gender and older age, particularly when associated, are known risk factors for bleeding complications in the case of the transfemoral approach,23 the radial artery should also, in our opinion, remain the preferred arterial access site in these cases. Our relatively high rate of crossover of 11.4% in the case of SVG interventions confirmed the anatomical and technical challenges of these interventions.24 Therefore, when a SVG is suspected to be the infarct-related artery, the transfemoral approach may represent a valuable alternative to TRA.

TRA–PPCI and procedural success

A systematic use of the approach, dedication of operators and staff, and high volumes are, in our opinion, the pillars for successful TRA, which is crucial in the setting of PPCI, because failure and delay in achieving arterial access could have a negative impact on reperfusion and procedural success rates. Although we did not systematically collect door-to-balloon times, which is independently associated with mortality,25 our procedural times include the first attempt at radial artery puncture and thereby are unlikely to be affected by a significant delay in obtaining arterial access. In addition, two observational studies have already showed comparable door-to-balloon times in both strategies of arterial access.26 27

Previously, procedural success rates of TRA–PPCI have been reported to range from 88% to 96%,4–7 14 which were not inferior to those achieved with the transfemoral approach. Most interventionalists are aware of the advantages of TRA in terms of vascular complications,15 20 21 but they are also afraid to achieve less optimal procedural success rates with TRA due to a delay in obtaining arterial access and/or technical limitations (less support, restricted choice of devices). As a consequence, TRA–PPCI still represents a minority of all emergency procedures performed worldwide.19

In our experience with more than 2000 unselected STEMI patients undergoing TRA–PPCI, procedural success rates remained high and stable throughout the whole study period (94.1%), despite a constant increase in procedural complexity (non-LAD infarct-related arteries, multivessel PPCI) and no restraints in the use of devices (thrombus aspiration, drug-eluting stenting). Furthermore, procedural and fluoroscopy times decreased substantially over the years, presumably because of both an increased proficiency of the operators and improvements in catheters and materials. The need for crossover had no apparent impact on procedural success rates. Although the procedural time was longer in the case of crossover, we presume that it depended not only on the delay needed to obtain vascular access but rather on a generally more complex patient or procedure as fluoroscopy time was equally longer in those patients.

Study limitations

Some limitations of this study deserve consideration. This analysis represents procedures performed in a high-volume centre and by interventional cardiologists highly devoted to TRA. The RIVAL study suggests that these findings would probably not apply to low-volume centres or operators practising TRA sporadically or as a bail-out approach. Furthermore, there could be unidentified factors and confounders, other than the demographic and clinical characteristics analysed, which influenced our results in terms of crossover and procedural performance. Finally, we lack information about vascular complications and clinical outcomes of our study cohort, because these parameters were not assessed in our cath-lab database. However, both TRA and procedural success have been extensively demonstrated to be potent surrogates of (the absence of) vascular complications and favourable clinical outcomes, respectively.8 21 28


The current analysis shows that a routine TRA for PPCI in an unselected cohort of STEMI patients is efficacious in achieving both high rates of arterial access and procedural success. The radial artery may thereby represent the arterial access site of choice for the majority of patients with STEMI undergoing PPCI.


The authors are grateful to the nursing staff of the catheterisation laboratory at the Onze Lieve Vrouwe Gasthuis for their contribution in preserving the completeness and accuracy of the data. They also thank Bob Springorum for his contribution to the maintenance of the institutional database.


View Abstract


  • Competing interests None.

  • Ethics approval The study was an observational study of data collected as part of standard patient care. Focus was on procedural aspects rather than patient outcomes.

  • Provenance and peer review Not commissioned; externally 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.