Author + information
- Received March 26, 2014
- Revision received May 6, 2014
- Accepted May 13, 2014
- Published online October 14, 2014.
- Mamas A. Mamas, DPhil, BM BCh∗,†,‡∗ (, )
- Simon G. Anderson, PhD, MB BCh∗,†,
- Matthew Carr, PhD§,
- Karim Ratib, MB BCh‖,
- Iain Buchan, MD‡,§,
- Alex Sirker, PhD, MB Chir¶,
- Douglas G. Fraser, MD, MB Chir∗,
- David Hildick-Smith, MD#,
- Mark de Belder, MD∗∗,
- Peter F. Ludman, MD††,
- James Nolan, MD‖,
- British Cardiovascular Intervention Society and the National Institute for Cardiovascular Outcomes Research
- ∗Manchester Heart Centre, Manchester Royal Infirmary, Manchester, United Kingdom
- †Cardiovascular Research Group, Institute of Cardiovascular Sciences, University of Manchester, Manchester, United Kingdom
- ‡Farr Institute, University of Manchester, Manchester, United Kingdom
- §Institute of Population Health, University of Manchester, Manchester, United Kingdom
- ‖University Hospital of North Staffordshire, Stoke-on-Trent, United Kingdom
- ¶The Heart Hospital, University College London Hospitals, London, United Kingdom
- #Sussex Cardiac Centre, Brighton and Sussex University Hospitals NHS Trust, Brighton, United Kingdom
- ∗∗The James Cook University Hospital, Middlesbrough, United Kingdom
- ††Queen Elizabeth Hospital, Birmingham, United Kingdom
- ↵∗Reprint requests and correspondence:
Dr. Mamas A. Mamas, Cardiovascular Institute, University of Manchester, Oxford Road, Manchester M13 9WL, United Kingdom.
Background Transradial access (TRA) has been associated with reduced access site–related bleeding complications and mortality after percutaneous coronary intervention (PCI). It is unclear, however, whether these observed benefits are influenced by baseline bleeding risk.
Objectives This study investigated the relationship between baseline bleeding risk, TRA utilization, and procedure-related outcomes in patients undergoing PCI enrolled in the British Cardiovascular Intervention Society database.
Methods Baseline bleeding risk was calculated by using modified Mehran bleeding risk scores in 348,689 PCI procedures performed between 2006 and 2011. Four categories for bleeding risk were defined for the modified Mehran risk score (MMRS): low (<10), moderate (10 to 14), high (15 to 19), and very high (≥20). The impact of baseline bleeding risk on 30-day mortality and its relationship with access site were studied.
Results TRA was independently associated with a 35% reduction in 30-day mortality risk (odds ratio [OR]: 0.65 [95% confidence interval (CI): 0.59 to 0.72]; p < 0.0001), with the magnitude of mortality reduction related to baseline bleeding risk (MMRS <10, OR: 0.73 [95% CI: 0.62 to 0.86]; MMRS ≥20, OR: 0.53 [95% CI: 0.47 to 0.61]). In patients with an MMRS <10, TRA was used in 71,771 (43.2%) of 166,083 PCI procedures; TRA was used in 8,655 (40.1%) of 21,559 PCI procedures in patients with an MMRS ≥20, illustrating that TRA was used less in those at highest risk from bleeding complications (p < 0.0001).
Conclusions TRA was independently associated with reduced 30-day mortality, and the magnitude of this effect was related to baseline bleeding risk; those at highest risk of bleeding complications gained the greatest benefit from adoption of TRA during PCI.
Advances in antithrombotic therapy have improved outcomes for patients undergoing percutaneous coronary intervention (PCI) by reducing ischemic events and mortality but with a corresponding increase in procedure-related bleeding complications. These peri-procedural bleeding complications are associated with adverse clinical outcomes, including myocardial infarction (MI), stroke, and death (1–3).
A significant proportion of such major bleeding complications are related to the access site. In a post-hoc analysis of the REPLACE-2 (Randomized Evaluation in PCI Linking Angiomax to Reduced Clinical Events), ACUITY (Acute Catheterization and Urgent Intervention Triage Strategy), and HORIZONS-AMI (Harmonizing Outcomes With Revascularization and Stents in Acute Myocardial Infarction) trials, 38.6% of major bleeds were access site related (4), and in the RIVAL (Radial vs. Femoral Access for Coronary Intervention) study, 30% of non–coronary artery bypass graft (CABG)-related major bleeds were access site related (5). A proportion of these access site bleeds correlate to the patients’ syndrome at presentation; patients with ST-segment elevation myocardial infarction (STEMI) exhibit a higher incidence of bleeding and a higher proportion of access site bleeds than those with non–ST-segment elevation myocardial infarction (NSTEMI) or unstable angina (6).
Transradial access (TRA) has been adopted as the default approach for PCI in many European and North American centers because of its proven beneficial effect on access site–related complications and mortality (6–8). It is widely believed that the mortality benefits seen with TRA are mediated through a reduction in major bleeding complications (9–11). A recent subgroup analysis of the RIVAL study found that TRA significantly reduces mortality as well as the primary composite outcome of mortality, MI, stroke, and non–CABG-related major bleeding in the STEMI subgroup; non-CABG major bleeding was not affected (12). In contrast, TRA was not associated with any impact on these outcomes in the NSTEMI cohort (12).
This finding may be related to the underlying bleeding risk in these patient groups. For example, 30-day Thrombolysis In Myocardial Infarction (TIMI) major bleeding increased from 0.7% in patients with stable angina to 1.6% and 2.6% in the NSTEMI and STEMI groups, respectively, in a recent patient-level pooled analysis of the REPLACE-2 (Randomized Evaluation of PCI Linking Angiomax to Reduced Clinical Events), ACUITY (Acute Catheterization and Urgent Intervention Triage strategY), and HORIZONS-AMI (Harmonizing Outcomes with Revascularization and Stents in Acute Myocardial Infarction) studies (13). Similarly, major bleeding increased from 1.1% in patients with stable angina to 1.8% and 4.5% in the NSTEMI and STEMI groups in an analysis of the National Cardiovascular Data Registry known as CathPCI (14). It is therefore possible that the greatest benefit of TRA occurs in those patients at greatest risk from bleeding complications, although previous research has not systematically studied the influence of baseline bleeding risk on access site–related outcomes.
The present paper is an observational analysis of TRA versus transfemoral access (TFA) utilization and their associated outcomes in a nonselected national cohort derived from the British Cardiovascular Intervention Society (BCIS) database over a 6-year period. We aimed to study the impact of baseline peri-procedural bleeding risk on access site–related outcomes and whether baseline bleeding risk influences access site choice within the United Kingdom.
The BCIS database
The BCIS collects data relating to the United Kingdom’s nationwide practice of PCI (15). The data are collected via the Central Cardiac Audit Database under the auspices of the National Institute for Cardiovascular Outcomes Research. The aim is to record all PCI procedures performed in any U.K. hospital. In 2011, ∼99% of all PCI procedures performed in U.K. National Health Service (NHS) hospitals had been entered into the database.
The BCIS database documents clinical, procedural, and outcomes information with a total of 113 variables. As of December 2011, there were ∼459,775 records in the BCIS database, with ∼80,000 new records added each year. Mortality tracking is performed by the Medical Research Information Service by using each patient’s NHS number, which provides a unique identifier for any person registered with the NHS in England and Wales.
The data presented relate to all reported PCI procedures undertaken in patients in the United Kingdom between January 1, 2006, and December 31, 2011. PCI procedures performed via the left or right femoral artery or the left or right radial artery were included in the TFA and TRA cohorts, respectively. Patients in whom the access site was unknown were excluded.
In-hospital major bleeding complications were defined as a composite of reported gastrointestinal bleed, intracerebral bleed, retroperitoneal hematoma, blood or platelet transfusion, or an arterial access site complication requiring surgery (7).
Two primary outcomes were examined. The first was 30-day mortality defined as any-cause death. The mortality data for each procedure were validated against Office for National Statistics records, the United Kingdom’s statutory deaths register. The second outcome was in-hospital major bleeding complication as previously defined.
Calculation of bleeding risk score
To predict non-CABG–related TIMI major bleeding in patients undergoing PCI in the elective and acute setting, the Mehran score was developed through a patient-level pooled analysis of the REPLACE-2, ACUITY, and HORIZONS-AMI trials (13). The risk score consists of 7 variables: serum creatinine level, age, sex, presentation, white blood cell count, cigarette smoking, and anticoagulant agent use.
In the present analysis, we modified the risk score published by Mehran et al. (13) to define the baseline risk of bleeding complications in patients who underwent PCI in the BCIS database. Our modified Mehran score, illustrated in Online Table 1, was calculated for each participant based on age, sex, clinical presentation, smoking history, renal function, and antithrombotic medications, with an integer score assigned for each category. Modifications were necessary because measurement of white blood cell count and actual serum creatinine, both of which are required to calculate the Mehran score, are not recorded in the BCIS database. Severe renal dysfunction was defined as a history of renal disease and creatinine levels >200 μmol/l or patients requiring dialysis. Moderate renal dysfunction was defined as a history of renal disease with creatinine levels <200 μmol/l or functioning transplant. No renal impairment was defined as no history of renal disease and creatinine levels <200 μmol/l as per the BCIS definition. The remaining categories and integer scores associated with each category remained unchanged from the original Mehran score (13).
All statistical calculations were performed by using Stata version 12.1 (Stata Corp, College Station, Texas). Data are expressed as arithmetic mean ± SD and number (%) for continuous and categorical variables, respectively. To compare proportions for nonparametric data, we used the chi-square test. Tests for linear trend used chi-square models.
Using the aforementioned template and as illustrated in Online Table 1, we calculated individual modified risk scores and further stratified procedures into 1 of 4 categories of risk bleeding as defined in Mehran’s original paper: low, moderate, high, and very high corresponding to integer scores <10, 10 to 14, 15 to 19, and ≥20.
We estimated odds ratios (ORs) of mortality or bleeding within 30 days per site of access (TFA vs. TRA) by using logistic regression models. Multivariate adjustment included the continuous variables age and dichotomized categorical variables (yes vs. no): sex (female vs. male), peripheral vascular disease, hypertension, hypercholesterolemia, type 2 diabetes, bivalirudin use, and glycoprotein IIb/IIIa inhibitor use. The other categorized variables included smoking status (current or ex-smoker vs. never), acute coronary syndrome type (STEMI or NSTEMI vs. elective/stable), left ventricular function (poor or moderate vs. good), and renal impairment (severe or moderate vs. normal). The 4 categories of risk bleeding (<10, 10 to 14, 15 to 19, and ≥20) were also included (<10 as comparator). Both the Akaike and Bayesian information criteria were used to assess model parsimony and goodness-of-fit.
For all models, we tested for interaction by adding terms for access site and the modified Mehran risk category. Tests for statistically significant interactions used the Wald test of the terms; if the interaction terms were not significant, they were excluded from the final models.
Model discrimination (i.e., the degree to which the prognostic score enables the discrimination between patients with a favorable or unfavorable outcome) was quantified by using the C-statistic, interpreted as the area under the receiver-operating characteristic curve for a corresponding logistic regression model (16). Bootstrapping techniques were used to validate the model (i.e., to adjust the estimated model performance for overoptimism or overfitting). We also adjusted our estimate of the C-statistic for optimism by using 10-fold cross-validation with random resampling to generate average predicted probabilities (17).
The Hosmer-Lemeshow goodness-of-fit test (18) for logistic regression models was used to assess the validity of the models. The statistic is computed as the Pearson chi-squared test from a contingency table of observed and expected frequencies. A significant test result indicates one important kind of invalidity of a model.
To control for potentially confounding factors, we used propensity score methods with nearest-neighbor matching to account for the confounding arising from differences in clinical characteristics of patients treated with TRA versus TFA. A propensity score was estimated for each episode of radial or femoral access given the following variables: age, sex, diabetes, hypertension, cholesterol, previous MI, CABG, syndrome of presentation, peripheral vascular disease, and modified Mehran score. Mahalanobis distance matching with the propensity score as the distance matrix was then undertaken to generate a 1:1 matched cohort. Standardized differences between groups were estimated to assess the balance achieved by matching; a baseline characteristic was considered well balanced if the standardized difference was <10%. We then estimated ORs of mortality or bleeding within 30 days in relation to the site of access (TFA vs. TRA) by using logistic regression models in the adjusted dataset.
Baseline clinical characteristics and incidence of peri-procedural factors
A total of 348,689 PCI procedures were performed in patients in the United Kingdom between January 1, 2006, and December 31, 2011, in which access site utilization was known and in which a modified Mehran score could be calculated (Figure 1). In 151,412 patients (43.4%), TRA was used, and TFA was used in 197,277 patients (56.6%). Patients undergoing PCI via TFA were older and more likely to be diabetic (Table 1). Those who underwent PCI via TRA were more likely to present with an acute coronary syndrome. Table 1 also details procedural characteristics. A greater proportion of PCI procedures via TFA were performed in patients with cardiogenic shock or requiring an intra-aortic balloon pump. Glycoprotein IIb/IIIa and drug-eluting stent usage was more common in procedures via TRA.
Association of modified Mehran bleeding risk score and in-hospital major bleeding
Baseline bleeding risk was determined for each of the 348,689 participants by calculating individual modified Mehran scores, and its relationship with in-hospital major bleeding events was studied. Figure 2 illustrates the relationship between crude major in-hospital bleeding rates in all 348,689 procedures and modified Mehran bleeding risk score category (13); low, moderate, high, and very high bleeding risk corresponded to integer scores <10, 10 to 14, 15 to 19, and ≥20, respectively. Thus, as the modified Mehran score increased, the risk of a significant bleed rose substantially, with each unit increase in the modified Mehran score associated with a 10% additional risk of in-hospital major bleed (OR: 1.10 [95% confidence interval (CI): 1.09 to 1.11]; p < 0.0001). Multivariate logistic regressions were performed to identify independent predictors of in-hospital major bleeding complications in the patients who underwent PCI. The naive estimate of the C-statistic for the model was 0.77 (0.75 to 0.78), and the adjusted C-statistic was 0.76 (0.75 to 0.78) after bootstrapping. We found no interaction between access site and the Mehran score when the interaction terms were included in the model. Among the strongest independent predictors of in-hospital major bleeding complications was baseline bleeding risk as assessed by using the Mehran score (Table 2).
Impact of bleeding risk on all-cause mortality
The impact of baseline bleeding risk on 30-day mortality outcomes was subsequently studied. Figure 3 illustrates the crude 30-day mortality in all 348,689 patients and its relationship with baseline bleeding risk as calculated by using the modified Mehran risk score. As before, increasing baseline bleeding risk (assessed according to the modified Mehran score) was associated with increased risk of 30-day mortality. Each unit increase in baseline bleeding risk (assessed according to the modified Mehran score) was associated with increased risk of 30-day mortality (univariate OR: 1.18 [95% CI: 1.17 to 1.18]; p < 0.0001); the adjusted C-statistic was 0.76 (0.76 to 0.77) after bootstrapping.
Multivariate logistic regression was performed to identify independent predictors of 30-day mortality in the patients undergoing PCI. The strongest independent predictor of 30-day mortality was bleeding risk at baseline (Table 2). For example, patients at very high bleeding risk (modified Mehran score ≥20) had an associated OR of 7.86 (95% CI: 6.61 to 9.33; p < 0.0001) for 30-day mortality; even those at moderate risk of bleeding complications (modified Mehran score 10 to 14) had an associated OR of 2.23 (95% CI: 1.89 to 2.63; p < 0.0001). Other independent predictors of 30-day mortality included diabetes (OR: 1.37 [95% CI: 1.21 to 1.54]; p < 0.0001) and TRA (OR: 0.82 [95% CI: 0.73 to 0.91]; p < 0.0001). The adjusted C-statistic for this model was 0.86 (0.85 to 0.88) after bootstrapping. No interaction was found between access site and Mehran score when the interaction terms were included in this model.
Relationship between bleeding risk and access site choice
We subsequently investigated whether baseline bleeding risk was associated with access site choice. The baseline clinical and procedural demographic characteristics in the 4 groups stratified according to baseline bleeding risk are presented in Table 3. At the lowest risk of bleeding complications (modified Mehran score <10), TRA was used in 71,771 (43.2%) of 166,083 PCI procedures; among the 21,599 patients at highest risk of bleeding complications (modified Mehran score ≥20), TRA was used in 8,655 (40.1%) of 21,559 patients, paradoxically illustrating that TRA was used significantly less as an access site in those at highest risk from bleeding complications (p < 0.0001).
Relationship between bleeding risk and access site choice on mortality and bleeding outcomes
Mortality outcomes were studied according to access site utilization. The 30-day mortality in the TRA cohort was 1,384 (0.91%) of 151,412 and 3,434 (1.74%) of 197,277 in the TFA cohort. After adjustment of baseline covariates, multivariate logistic regression analysis revealed that TRA was independently associated with a reduction in 30-day mortality (OR: 0.65 [95% CI: 0.59 to 0.72]; p < 0.0001).
The impact of baseline bleeding risk on 30-day mortality according to access site was also studied. Figure 4A illustrates the actual 30-day mortality rates in the TRA and TFA cohorts. As before, as the baseline risk of bleeding (calculated with the modified Mehran score) rises, both cohorts experienced an increase in 30-day mortality. However, there was a clear separation in 30-day mortality for the TRA and TFA cohorts as baseline bleeding risk increased; the absolute difference in 30-day mortality was thus greatest at the highest baseline bleeding risk (as defined by using the modified Mehran score). A similar relationship was observed for major bleeding events (Figure 4B).
Figure 5 illustrates the ORs for 30-day mortality for TRA versus TFA access according to the 4 modified Mehran categories of bleeding risk. The mortality benefit associated with TRA increased as the bleeding risk (defined according to the modified Mehran score) increased. For example, in the lowest bleeding risk category (modified Mehran score <10), TRA was associated with a 27% reduction in 30-day mortality (OR: 0.73 [95% CI: 0.62 to 0.86]); in the highest bleeding risk category (modified Mehran score ≥20), TRA was associated with a 47% reduction in 30-day mortality (OR: 0.53 [95% CI: 0.47 to 0.61]), suggesting that the greatest benefit of TRA was recorded in those at highest risk of bleeding complications.
Propensity score analyses
To account for confounding variables and bias in our study cohort, propensity score matching was performed to adjust for differences in clinical baseline variables (age, sex, diabetes, hypertension, hypercholesterolemia, peripheral vascular disease, previous PCI, syndrome of presentation, previous CABG, and the modified Mehran risk score and their interaction terms), producing a total of 102,664 matched sets (radial vs. femoral access). The baseline demographic characteristics of the propensity-matched cohorts stratified according to access site (Online Table 2) and access site data according to modified Mehran scores of <10, 10 to 14, 15 to 19, and ≥20 (Online Tables 3 to 6) are presented.
The crude rates for mortality increased in both the TRA and TFA cohorts as risk of bleeding calculated with modified Mehran score increased. The absolute difference in 30-day mortality rose as the baseline bleeding risk (defined by using the modified Mehran score) increased. For example, 30-day mortality at lowest bleeding risk (Mehran score <10) was 0.46% and 0.32% (p = 0.0005), 1.27% and 0.85% (p < 0.0001) for a risk score of 10 to 14, and 4.01% and 2.00% (p < 0.001) with a risk of 15 to 19 for the TFA and TRA propensity-matched groups, respectively. In the highest baseline bleeding risk group (Mehran score ≥20), 30-day mortality was 6.53% and 3.10% in the TFA and TRA groups, respectively (p < 0.0001). The ORs for 30-day mortality are presented in Table 4, which shows that the mortality benefit associated with TRA increased as the bleeding risk (defined according to the modified Mehran score) increased.
In our current analysis of nearly 350,000 PCI procedures performed in the United Kingdom, we found that use of TRA was independently associated with a reduction in 30-day mortality outcomes and that the magnitude of this mortality effect was related to baseline bleeding risk, with those at highest risk of bleeding complications gaining the greatest benefit from adoption of TRA for PCI. Furthermore, our analysis suggests that in a national setting, access site choice does not seem to be influenced by baseline bleeding risk and, paradoxically, that TRA utilization rates are lower in those patients at highest risk of bleeding complications.
Peri-procedural major bleeding complications are among the most powerful independent predictors of mortality (1–3,13,19,20), with a greater impact on mortality than peri-procedural MI (13,20). Our data also suggest that baseline bleeding risk is one of the strongest predictors of 30-day mortality for those patients at highest baseline risk from bleeding complications (modified Mehran score ≥20); they had an 8-fold greater risk of 30-day mortality than those at lowest risk from bleeding complications (modified Mehran score <10).
Several studies have suggested that at least one-third of such bleeding complications are related to access site (4,5), with TFA a major determinant of adverse outcomes in those patients who sustain bleeding complications during PCI (21). Adoption of TRA has been shown to be associated with lower mortality (6–8) through a reduction in major bleeding complications in various studies (9–11).
Data derived from previous studies suggest that the greatest benefit of TRA in reducing mortality outcomes may be achieved in those at greatest risk of bleeding complications. A recent subgroup analysis of the RIVAL study found that TRA reduced mortality as well as the primary composite outcome in the STEMI subgroup but was not associated with any impact on these outcomes among the NSTEMI subgroup (12). This finding may have been related to the higher baseline bleeding risk of the STEMI versus NSTEMI cohort. For example, in RIVAL, non-CABG–related major bleeding occurred in 0.87% of the STEMI cohort and 0.57% of the NSTEMI group, whereas major bleeding in the ACUITY study occurred in 3.1% of the STEMI group and 2.26% of the NSTEMI group.
To our knowledge, our analysis is the first to investigate the influence of baseline bleeding risk on access site–related outcomes. We found a separation in mortality risk between TRA and TFA increasing as baseline bleeding risk increased, with the maximal beneficial effect of TRA observed in those patients with the greatest baseline bleeding risk. However, no association between baseline bleeding risk and access site choice was observed.
Various possible explanations may account for our observations. First, many of the bleeding risk stratification scores have been published recently and therefore could not have been used before 2010. Second, many of these risk scores require blood results such as serum creatinine (13,20,22), white blood cell count (13,20), baseline hemoglobin (20), or hematocrit (22), which may not be available on presentation in many patients at the highest risk of bleeding complications (e.g., those undergoing primary PCI or patients with cardiogenic shock or unstable NSTEMI); this limitation makes assessment of baseline bleeding risk using these risk stratification scores difficult.
Third, procedural-related factors might also affect access site choice in patients at highest risk of bleeding complications. Both female sex and increasing age raise bleeding risk (13,20,22,23), but they also make TRA more technically challenging. Finally, most of the risk stratification scores have an integer score assigned to patients presenting with STEMI (13,20,23) who are at highest risk of bleeding complications. Given that such patients may be unstable at presentation, operators (particularly at the start of their learning curve) might find TRA challenging and therefore opt for TFA instead.
The BCIS dataset includes an almost complete collection of all PCI procedures performed in the United Kingdom. In addition to incorporating all-comers, this dataset reflects a national, real-world experience that includes many high-risk patients encountered daily who are often excluded from randomized controlled trials. Furthermore, the present analysis included close to 350,000 PCI procedures, representing one of the largest analyses of access site–related outcomes in PCI and the first to systematically study the relationship between those outcomes and baseline bleeding risk.
First, only 80% of the cohort contained in the BCIS dataset (through December 31, 2011) was included in this analysis; the remaining 20% had data missing for ≥1 parameter necessary to calculate our modified Mehran score. Second, we modified the risk scores published by Mehran to define the baseline risk of bleeding complications in BCIS database patients who underwent PCI. Modifications were necessary because actual serum creatinine or white blood cell count information included in the Mehran score is not entered into the BCIS database, although other score components remained unchanged. Although our predicted rates of bleeding for a given Mehran score are lower than the original Mehran score report (13), this finding may be related to both differences in the definition of major bleeding used in the current analysis as well as the omission of white blood cell count from our modified score. Nevertheless, our modified Mehran score demonstrated a strong relationship between increasing integer score and the escalating risk of major in-hospital bleeding complications, with each unit increase in the modified Mehran score associated with a 10% increased risk of in-hospital major bleeding. The modified Mehran risk score was the strongest predictor of in-hospital major bleeding complications on multivariate analysis.
Third, although mortality tracking within the United Kingdom is very robust, cause of death is not currently available, and all other outcomes and complications such as bleeding complications are self-reported and not formally validated by BCIS, potentially resulting in underreporting. Finally, bleeding definitions used in the BCIS dataset are different from those frequently used in clinical trials, such as ACUITY and TIMI; thus, we could not assess the performance of our modified Mehran score by using these bleeding definitions.
Use of TRA was independently associated with a reduction in 30-day mortality, and the magnitude of this mortality effect was related to baseline bleeding risk: those at highest risk of bleeding complications gained the greatest benefit from TRA adoption for PCI. Our analysis suggests that in a national setting, access site choice does not currently seem to be influenced by baseline bleeding risk and that use of TRA was lowest in those patients at highest risk of bleeding complications.
Our data support the formal assessment of bleeding risk as part of the decision-making process for optimal access site choice before PCI, in keeping with the current American College of Cardiology Foundation/American Heart Association/Society for Cardiovascular Angiography and Interventions guidelines for PCI in which evaluation of bleeding risk before PCI was given a class I recommendation (24). Combining TRA with optimal pharmacological strategies has great potential to reduce both access site and systemic bleeding, resulting in better patient outcomes after PCI, particularly in the groups of patients at highest risk of bleeding complications, such as those presented in the Central Illustration.
COMPETENCY IN PATIENT CARE: The patient’s risk of bleeding should be carefully considered in selecting the optimal approach to vascular access in patients undergoing percutaneous coronary intervention.
COMPETENCY IN MEDICAL KNOWLEDGE: Patients at highest risk of bleeding complications gain the greatest benefit from use of radial artery access for percutaneous coronary intervention.
TRANSLATIONAL OUTLOOK: Validation of risk assessment schemes that encompass a broad array of clinical outcomes, including the success of revascularization as well as bleeding and other complications to guide optimal selection of vascular access sites, could strengthen the evidence base and accelerate their incorporation into practice guidelines.
Dr. de Belder has reported receiving a travel grant from Medtronic. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose. Drs. Mamas and Anderson contributed equally to this work.
- Abbreviations and Acronyms
- British Cardiovascular Intervention Society
- coronary artery bypass graft
- confidence interval
- myocardial infarction
- National Health Service
- non–ST-segment elevation myocardial infarction
- odds ratio
- percutaneous coronary intervention
- ST-elevation myocardial infarction
- transfemoral access
- Thrombolysis In Myocardial Infarction
- transradial access
- Received March 26, 2014.
- Revision received May 6, 2014.
- Accepted May 13, 2014.
- American College of Cardiology Foundation
- Eikelboom J.W.,
- Mehta S.R.,
- Anand S.S.,
- Xie C.,
- Fox K.A.,
- Yusuf S.
- Manoukian S.V.,
- Feit F.,
- Mehran R.,
- et al.
- Rao S.V.,
- Cohen M.G.,
- Kandzari D.E.,
- Bertrand O.F.,
- Gilchrist I.C.
- Mamas M.A.,
- Ratib K.,
- Routledge H.,
- et al.,
- on behalf of the British Cardiovascular Intervention Society and the National Institute for Cardiovascular Outcomes Research
- Rao S.V.,
- Ou F.S.,
- Wang T.Y.,
- et al.
- Chase A.J.,
- Fretz E.B.,
- Warburton W.P.,
- et al.
- Mehta S.R.,
- Jolly S.S.,
- Cairns J.,
- et al.,
- for the RIVAL Investigators
- Mehran R.,
- Pocock S.,
- Nikolsky E.,
- et al.
- Subherwal S.,
- Peterson E.D.,
- Dai D.,
- et al.
- Ludman P.F.
- Hosmer D.W.,
- Lemeshow S.,
- Sturdivant R.X.
- Mehran R.,
- Pocock S.J.,
- Stone G.W.,
- et al.
- Mehran R.,
- Pocock S.J.,
- Nikolsky E.,
- et al.
- Doyle B.J.,
- Ting H.H.,
- Bell M.R.,
- et al.
- Subherwal S.,
- Bach R.G.,
- Chen A.Y.,
- et al.
- Mehta S.K.,
- Frutkin A.D.,
- Lindsey J.B.,
- et al.
- Levine G.N.,
- Bates E.R.,
- Blankenship J.C.,
- et al.,
- American College of Cardiology Foundation; American Heart Association Task Force on Practice Guidelines; Society for Cardiovascular Angiography and Interventions