Author + information
- Received March 27, 2013
- Revision received October 4, 2013
- Accepted October 7, 2013
- Published online March 1, 2014.
- Edward L. Hannan, PhD∗∗ (, )
- Louise Szypulski Farrell, MS∗,
- Gary Walford, MD†,
- Peter B. Berger, MD‡,
- Nicholas J. Stamato, MD§,
- Ferdinand J. Venditti, MD‖,
- Alice K. Jacobs, MD¶,
- David R. Holmes Jr., MD#,
- Samin Sharma, MD∗∗ and
- Spencer B. King III, MD††
- ∗University at Albany, State University of New York, Albany, New York
- †Johns Hopkins University, Baltimore, Maryland
- ‡Geisinger Medical Center, Danville, Pennsylvania
- §United Health Services, Binghamton, New York
- ‖Albany Medical Center, Albany, New York
- ¶Boston Medical Center, Boston, Massachusetts
- #Mayo Clinic, Rochester, Minnesota
- ∗∗Mount Sinai Hospital, New York, New York
- ††St. Joseph's Health System, Atlanta, Georgia
- ↵∗Reprint requests and correspondence:
Dr. Edward L. Hannan, School of Public Health, State University of New York, University at Albany, One University Place, Rensselaer, New York 12144-3456.
Objectives This study sought to determine the utilization and outcomes for radial access for percutaneous coronary intervention (PCI) for ST-segment elevation acute myocardial infarction (STEMI) in common practice.
Background Radial access for PCI has been studied considerably, but mostly in clinical trials.
Methods All patients undergoing PCI for STEMI in 2009 to 2010 in New York were studied to determine the frequency and the patient-level predictors of radial access. Differences in in-hospital/30-day mortality between radial and femoral access were also studied.
Results Radial access increased from 4.9% in the first quarter of 2009 to 11.9% in the last quarter of 2010. Significant independent predictors were higher body surface area, non-Hispanic ethnicity, Caucasian race, stable hemodynamic state, ejection fraction <30% and ≥50% onset of STEMI from 12 to 23 h before the index procedure, and peripheral vascular disease. Mortality was not related to access site after adjustment for covariates (for radial vs. femoral access, adjusted odds ratio: 0.86, 95% confidence interval: 0.59 to 1.25), but the radial access site was trending toward lower mortality for the 9 hospitals that used it for more than 10% of their patients (adjusted odds ratio: 0.61, 95% confidence interval: 0.36 to 1.02).
Conclusions The use of a radial access site for PCI in STEMI patients increased between 2009 and 2010, but was still infrequent in 2010, and was used for lower-risk STEMI patients. There was no significant difference in mortality by access site, but there was a trend toward a mortality advantage for patients with a radial access site among hospitals that used it relatively frequently.
In recent years, numerous studies have shown that the radial artery is a safe and feasible alternative to the femoral artery as a vascular access site for percutaneous coronary intervention (PCI) and coronary angiography. Among the advantages of radial artery access are lower bleeding and vascular complication rates and earlier discharge because there is no requirement for post-procedural bed rest (1–3). This has also resulted in lower costs (4). However, the transradial approach to PCI is more technically demanding, and its adoption in the United States has been slow despite the fact that it is more commonly used in Asia and Europe (4,5). More recently, numerous clinical trials and observational studies have found superior outcomes for the radial approach, particularly for patients presenting with an acute myocardial infarction (AMI) and for the subset of AMI patients presenting with ST-segment elevation myocardial infarction (STEMI) (1–21).
One purpose of this study was to examine the utilization of the radial approach for patients undergoing PCI for STEMI in New York State, including trends and variations across hospitals in the utilization of the radial versus femoral approaches. The study also identifies patient factors that are associated with higher use of the radial approach, compares short-term (in-hospital/30-day) mortality after adjusting for important differences between the 2 groups, and examines differences in door-to-balloon times.
Data were obtained from New York State's Percutaneous Coronary Interventions Reporting System, a mandatory registry in New York that was initially developed in 1991. The registry contains detailed information about each patient undergoing PCI in the state, including demographics; pre-procedural risk factors; periprocedural complications; types of devices used; extent of disease and lesions treated; dates of admission, discharge, and procedure; discharge disposition and destination; and hospital and operator identifiers. There is also a variable that denotes the access site (radial, femoral).
These data are matched to New York State administrative data for purposes of auditing completeness and the accuracy of in-hospital mortality reporting. Also, medical records are audited by the New York State Department of Health's utilization review agent in order to ensure accuracy of the reporting of risk factors and complications. Registry records are matched to New York vital statistics data to identify deaths after discharge within 30 days of the index procedure.
Patients, hospitals, and outcomes
A total of 11,057 STEMI patients with an onset-to-door time of <24 h who underwent immediate primary or rescue PCI in nonfederal New York hospitals between January 1, 2009, and December 31, 2010, were considered for the study. A total of 840 (7.6%) patients had radial access, 7 of whom had radial access followed by femoral access (and were counted as radial access), and 10,217 (92.4%) of whom had femoral access. Patients were excluded if they had cardiogenic shock before or at the time of the PCI.
All 67 hospitals in New York State in which PCI was performed for STEMI were included in the study. The primary outcomes studied were the use of a radial access site and short-term mortality (in-hospital/30-day).
All analyses were conducted for the group of STEMI patients mentioned in the preceding text. First, the prevalence and percentage of patients undergoing PCI with a radial access site were tracked in 3-month intervals throughout the course of the study to identify the relative frequency of radial access and to determine whether the extent to which radial access was chosen increased substantially during the course of the study period. Variation in the use of radial access across hospitals was computed after subdividing the percentage of patients with radial access into relevant ranges.
The significant bivariate predictors of radial access were identified based on available information in the registry, which included demographics (age, sex, race, ethnicity), hemodynamic state, ejection fraction, time from onset of pre-procedural myocardial infarction, Canadian Cardiovascular Society class, type of angina, and numerous comorbidities. Chi-square tests were used for categorical variables, and Wilcoxon rank sum tests were used for continuous variables to determine whether there was a significantly higher prevalence of 1 type of access site for each variable in comparison to the other type of access site. Variables that were associated with a higher prevalence of 1 of the types of access sites when in competition with all other variables (i.e., significant independent predictors of access site) were identified by conducting a stepwise logistic regression analysis with access site as the dependent variable, and all significant bivariate predictors (with p < 0.10) of access site as candidate independent variables.
Bivariate differences in in-hospital/30-day mortality between radial and femoral access were examined along with the bivariate relationships with mortality of all of the risk factors mentioned in the preceding text. Significant risk factors in the bivariate analyses (p < 0.10) were added to access site in a stepwise logistic regression analysis with generalized estimating equations to determine whether access site was a significant predictor of mortality after controlling for the impact of significant patient-level risk factors. Generalized estimating equations were used to account for clustering of patients within hospitals. Differences in door-to-balloon times were also investigated.
All tests were 2-sided and conducted at the 0.05 level, and all analyses were conducted in SAS version 9.1 (SAS Institute, Cary, North Carolina).
The percentage of all STEMI patients undergoing PCI with a radial access site rose monotonically during the course of the study, from 4.9% in the first quarter of 2009 to 11.9% during the last quarter of 2010 (Fig. 1).
The variation across hospitals in the use of a radial access site during the study period was considerable (Fig. 2). Twelve (21%) of the 58 hospitals never used a radial access site, 44 (76%) of the hospitals used a radial access site <5% of the time, and 4 (7%) of the hospitals used radial access for >30% of their PCIs. The post-procedural length of stay for STEMI patients with radial access was significantly lower (3.6 days vs. 4.6 days, p < 0.0001).
Bivariate predictors of radial access for PCI for STEMI patients were higher body surface area, non-Hispanic ethnicity, race, stable hemodynamic state, higher and lower ejection fractions, STEMIs with onset times from 6 to 23 h, peripheral vascular disease, absence of malignant ventricular arrhythmia, chronic obstructive pulmonary disease, low creatinine levels, absence of multivessel disease, no previous PCI or coronary artery bypass graft surgery, no previous organ transplant, urgent priority, Canadian Cardiovascular Society classes I, III, and IV, and unstable angina (Table 1). Significant independent predictors of radial access for STEMI patients undergoing PCI in the logistic regression model were higher body surface area, non-Hispanic ethnicity, Caucasian race, stable hemodynamic state, ejection fraction <30% and ≥50%, onset of STEMI between 12 and 23 h before the index procedure, and peripheral vascular disease (Table 2).
With regard to mortality, in-hospital/30-day mortality for all STEMI patients with radial access PCI was lower, but not significantly lower, than the mortality for their femoral access counterparts (2.7% vs. 3.6%; p = 0.21). Significant independent predictors of in-hospital/30-day mortality for STEMI patients were older age, hemodynamic instability, ejection fraction <30%, peripheral vascular disease, malignant ventricular arrhythmia, chronic obstructive pulmonary disease, current congestive heart failure, elevated creatinine levels, renal dialysis, multivessel disease, and left main disease. After controlling for these risk factors, there was not a significant mortality difference by access site (for radial vs. femoral access, adjusted odds ratio [AOR]: 0.86, 95% confidence interval [CI]: 0.59 to 1.25).
When the analyses were restricted to the 9 hospitals that used radial access for ≥10% of their STEMI patients, radial access was trending toward lower mortality in comparison to femoral access in those 9 hospitals (AOR: 0.61, 95% CI: 0.36 to 1.02). These 9 hospitals used PCI with radial access for 660 patients in total, and this constituted 31% of all PCIs for their STEMI patients and 79% of all PCIs for STEMI with radial access that were performed in the state in the study period.
In the group of all 58 hospitals in the study, the respective percentages of STEMI patients undergoing PCI with radial or femoral access with door-to-balloon times <90 min were 72% and 77% (p = 0.0006).
The purpose of this study was to examine the use of radial artery access for STEMI patients undergoing PCI in New York in 2009 to 2010, including changes in use, patient characteristics associated with radial access, and differences in mortality between patients receiving radial and those receiving femoral access.
With regard to utilization patterns, we found that there was a substantial increase in the use of radial artery access for PCI among STEMI patients (from 4.9% in the first quarter of 2009 to 11.9% in the last quarter of 2010). However, as evidenced by these percentages, the conversion to radial access is still quite modest, although it is somewhat higher in New York than that reported in the United States based on the NCDR (National Cardiovascular Data Registry) (6.4% of STEMI patients received radial access in the third quarter of 2011) (19). Clearly, disadvantages associated with radial access (a longer learning curve and more technically demanding) are still important considerations despite the increase in its use.
It is also notable that there is considerable variation in choice of arterial access site for AMI patients across hospitals in the state. Although 76% of all hospitals used radial access for AMI patients <5% of the time, and 21% of hospitals never used it, there were 4 hospitals that used radial access for >30% of their patients. Whether this is a sign of a trend toward radial access across the state remains to be seen.
The significant independent correlates of radial access for STEMI patients were almost all related to lower risk and nonminority status: higher body surface area, Caucasian race, non-Hispanic ethnicity, hemodynamic instability, higher and lower ejection fractions, STEMI within 12 to 23 h before the procedure, and peripheral vascular disease. In comparison, Baklanov et al. (21) found younger age, male sex, white race, peripheral vascular disease, recent heart failure, and no prior AMI, PCI, coronary artery bypass graft surgery, or dialysis to be associated with higher radial access rates.
Thus, although our study pertained to high-risk patients (STEMI), the subset of these patients undergoing radial access was of lower risk than the group undergoing femoral access. As a result of this, although the in-hospital/30-day mortality rate for STEMI patients undergoing PCI with radial access was somewhat lower than the rate for their femoral access counterparts (2.7% vs. 3.6%; p = 0.21), the AOR for mortality of radial versus femoral access was not significant (AOR: 0.86, 95% CI: 0.59 to 1.25). However, when the analyses were restricted to the 9 hospitals that used radial access for ≥10% of their AMI patients, radial access was trending toward lower mortality (AOR: 0.61, 95% CI: 0.36 to 1.02). Also, as noted earlier, there is no reason to believe that these hospitals had superior outcomes in general compared with other hospitals because their overall risk-adjusted mortality was not lower.
It should be noted that several randomized controlled trials (RCTs) and observational studies found lower adverse outcome rates for AMI patients undergoing PCI with radial access. For example, in a meta-analysis of 9 RCTs consisting of 2,977 STEMI patients undergoing PCI, Mamas et al. (6) found that patients with radial access had significantly lower rates of mortality (odds ratio [OR]: 0.53, 95% CI: 0.33 to 0.84), major adverse cardiac events (OR: 0.62, 95% CI: 0.43 to 0.90), and access site complications (OR: 0.30, 95% CI: 0.19 to 0.48). Joyal et al. (7), in a meta-analysis of 10 RCTs of primary PCI for STEMI with 3,347 patients, found that the radial approach was associated with better survival (OR: 0.53, 95% CI: 0.33 to 0.84), and a lower rate of vascular complication/hematoma (OR: 0.35, 95% CI: 0.24 to 0.53). Jang et al. (1) performed a meta-analysis of 21 RCTs and observational studies for STEMI patients undergoing PCI, and found that radial access was associated with lower rates of major adverse cardiac events (OR: 0.56, 95% CI: 0.44 to 0.72), mortality (OR: 0.55, 95% CI: 0.42 to 0.72), and major bleeding (OR: 0.32, 95% CI: 0.22 to 0.48).
A large observational study by Baklanov et al. (21) using NCDR data limited to STEMI patients found that patients with radial access had significantly lower in-hospital mortality than patients with femoral access (AOR: 0.76, 95% CI: 0.57 to 0.99; p = 0.0455). The significant findings may have in part been related to the mortality measure used in the study. When we repeated our study using only in-hospital mortality rather than in-hospital/30-day mortality, we also found that the radial access had significantly lower risk-adjusted mortality. However, we believe in-hospital/30-day mortality is a better measure to use because of relatively high short-term mortality rates for PCI patients after discharge. Also, our study differed from the NCDR study in that our study is population-based (it includes all patients undergoing PCI in a large geographical region).
An observational study conducted in Italy by Valgimigli et al. (14) found that radial access PCI was associated with lower 2-year risk-adjusted mortality for STEMI patients than femoral access PCI (8.8% vs. 11.4%; p = 0.03). However, it should be noted that 28% of all patients in the study underwent PCI with radial access, so their findings may be stronger in favor of radial access than ours because of increased experience using the radial approach. When we limited our study to hospitals using radial access for ≥10% of their STEMI patients, we found that radial access was trending toward significantly lower in-hospital/30-day mortality. This corresponds to the finding of Mehta et al. (16) that there was a correlation between hospitals' increased use of radial access and better outcomes.
The primary study limitation is the use of an observational database in order to compare mortality for patients with radial and femoral access. This can lead to the chance of unmeasured confounding because of variables not available in our database that may have been a factor in the choice of arterial access that were also related to in-hospital/30-day mortality. Typical ways of minimizing this bias include propensity matching of patients and controlling for variables available in the observational database using multivariable statistical models. We chose the latter approach via a logistic regression model. Nevertheless, unmeasured confounding is possible, and may have contributed to our finding that although radial access was associated with lower mortality than femoral access, it was not significantly lower. Another limitation of our study is that we did not have access to data on bleeding, which has been demonstrated to have lower rates among radial access patients (21).
Also, only 7 patients were reported as crossovers from radial access to femoral access, and because our analyses were based on intention to treat, and crossover probably has worse outcomes on average, unreported crossovers will bias the results against femoral access patients.
The use of a radial access site for PCI in STEMI patients increased between 2009 and 2010, but was still infrequent in 2010, and was used for lower-risk STEMI patients. There was no significant difference in mortality by access site, but there was a trend toward a mortality advantage for patients with a radial access site among hospitals that used it relatively frequently.
The authors would like to thank New York State's Cardiac Advisory Committee (CAC) for their encouragement and support of this study; and Kimberly S. Cozzens and Cynthia Johnson and the cardiac catheterization laboratories of the participating hospitals for their tireless efforts to ensure the timeliness, completeness, and accuracy of the registry data.
This study was funded in part by the New York State Department of Health. Dr. Berger has received consultant fees from Medicure and Janssen Pharmaceuticals; and funding as an investigator for the Geisinger Clinic from AstraZeneca, The Medicines Company, Bristol-Myers Squibb, sanofi-aventis, and Eli Lilly and Company/Daiichi Sankyo. Dr. Jacobs is Site PI on the Xience V Everolimus Eluting Coronary Stent System (EECSS) USA Post-Approval Study. Dr. Sharma is on the Speaker's Bureau of Boston Scientific, Abbott Vascular, Angioscore, Eli Lilly and Company/Daiichi Sankyo, and The Medicines Company. Dr. King has received consulting fees from Merck & Company, Wyeth Pharmaceuticals, nContact Surgical, Medtronic, and Celonova Biosciences; and is the editor-in-chief of JACC: Cardiovascular Interventions. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.
- Abbreviations and Acronyms
- acute myocardial infarction
- adjusted odds ratio
- confidence interval
- odds ratio
- percutaneous coronary intervention
- randomized controlled trial
- ST-segment elevation myocardial infarction
- Received March 27, 2013.
- Revision received October 4, 2013.
- Accepted October 7, 2013.
- American College of Cardiology Foundation
- Romagnoli E.,
- Biondi-Zoccai G.,
- Sciahbasi A.,
- et al.
- Mitchell M.D.,
- Hong J.A.,
- Lee B.Y.,
- Umscheid C.A.,
- Bartsch S.M.,
- Don C.W.
- Nadarasa K.,
- Robertson M.C.,
- Wong C.-K.,
- et al.
- Mamas M.A.,
- Ratib K.,
- Routledge H.,
- et al.
- Bertrand O.F.,
- Patel T.
- Généreux P.,
- Mehran R.,
- Palmerini T.,
- et al.,
- for the HORIZONS-AMI Trial Investigators
- Valgimigli M.,
- Saia F.,
- Guastaroba P.,
- et al.,
- REAL Registry Investigators
- Chodór P.,
- Kurek T.,
- Kowalczuk A.,
- et al.
- Mehta S.R.,
- Jolly S.S.,
- Cairns J.,
- et al.,
- RIVAL Investigators
- Jolly S.S.,
- Niemelä K.,
- Xavier D.,
- et al.
- Chase A.J.,
- Fretz E.B.,
- Warburton W.P.,
- et al.
- Brueck M.,
- Bandorski D.,
- Kramer W.,
- Wieczorek M.,
- Höltgen R.,
- Tillmanns H.
- Baklanov D.V.,
- Kaltenbach M.S.,
- Marso S.P.,
- et al.