Despite increasing use of the transradial approach (TRA) for coronary angiography, TRA failure and subsequent access site crossover remain a barrier to TRA adoption. The aim of this study was to elucidate patient and procedural characteristics associated with TRA to transfemoral approach (TFA) crossover and examine TRA to TFA crossover by operator experience over time. This retrospective analysis identified 1,600 patients who underwent coronary angiography with possible percutaneous coronary intervention through TRA by operators with varied TRA experience in an urban tertiary care center from October 2010 to August 2013. Univariate and multivariable logistic regression were used to identify independent predictors of access site crossover, from TRA to TFA, and strength of association is presented as odds ratio (OR, 95% confidence interval [CI]). Access site crossover was noted in 166 patients (10.4%). Multivariable predictors of access site crossover included age >75 years (OR 1.90, 95% CI 1.23 to 2.91, p = 0.004) and operator experience (OR 2.98, 95% CI 1.96 to 4.52, p <0.0001). Less experienced operators (≤5 years TRA experience) had a decrease in access site crossover over time (quartile 1: 8.9%, quartile 2: 18.8%, quartile 3: 16.4%, and quartile 4: 8.6%, p <0.001), which paralleled an increase in the proportion of procedures using initial TRA over time (quartile 1: 38.0%, quartile 2: 53.7%, quartile 3: 54.8%, and quartile 4: 70.3%, p <0.001). Experienced operators (>5 years TRA experience) had no significant change in proportion of access site crossover over time (quartile 1: 2.8%, quartile 2: 6.4%, quartile 3: 5.6%, quartile 4: 5.8%, p = 0.54). In conclusion, rate of access site crossover in the contemporary era is relatively low and can be mitigated with operator experience.
A transradial approach (TRA) to coronary angiography and percutaneous coronary intervention (PCI) is associated with decreased bleeding and access site complications, shorter hospital stays, early ambulation, and improved patient comfort compared with the transfemoral approach (TFA). TRA is also associated with decreased mortality in patients presenting with ST-segment elevation myocardial infarction. Although utilization of TRA is increasing in the United States, the overall prevalence still remains <20% of all procedures. Anatomical factors that decrease operator adoption and increase the learning curve associated with TRA include access site failure, radial artery spasm, radial and brachial loops, and tortuosity of the innominate trunk. These factors may also lead to increased rate of access site crossover. Access site crossovers can be potentially problematic as it increases procedure duration, radiation exposure, and risk of vascular complications related to multiple sites of access in patients on antiplatelet and antithrombotic therapy. This study aims to detail patient and procedural characteristics associated with access site crossover, from TRA to TFA. This study also aims to examine TRA to TFA crossover by operator experience as utilization of TRA increases over time.
Methods
This is a retrospective study of consecutive patients who underwent a diagnostic coronary angiography with or without PCI using TRA at a tertiary care center from October 2010 to August 2013. Patients who underwent a planned PCI without a diagnostic component were not included. For patients who had >1 procedure using TRA during the study period, only the first chronological procedure was selected. During this study period, transition to the opposite radial or ulnar artery was not routinely performed (n = 5) and, therefore, excluded from the study. This study was approved by the Institutional Review Board at the New York University School of Medicine and Bellevue Hospital Center.
Approach to coronary angiography and PCI, including access site and type of catheters used, was per individual operator practice. However, during the study period, standard practice was as follows. Conscious sedation was administered before local anesthesia with subcutaneous 1% lidocaine. Arterial access was obtained at least 2 cm proximal to the radial styloid process using the Seldinger technique. Once arterial access was obtained, a 5Fr or 6Fr hydrophilic sheath was inserted, and a cocktail of nitroglycerin 100 μg, verapamil 2.5 mg, and 2,500 U of unfractionated heparin was administered; 5Fr diagnostic catheters were used to cannulate the coronary arteries, and radial artery angiography was not performed unless the initial wire or catheter could not be passed to the ascending aorta.
Baseline demographic, clinical, and procedural characteristics were abstracted from a review of electronic patient medical records, including cardiac catheterization reports. Operator experience was categorized as those with less (≤5 years) or more (>5 years) TRA experience. The primary outcome of interest was access site crossover defined as crossover from TRA to TFA for any reason (access failure or failure to engage the coronary arteries or perform PCI).
Continuous variables were examined for normality using the Shapiro-Wilkes test. Differences in baseline characteristics between patients who did versus those who did not undergo access site crossover were evaluated by independent 2-sample t tests for normally distributed continuous variables, Mann–Whitney U test for skewed continuous variables, and test of proportions for categorical variables. Logistic regression was used to determine multivariable predictors of access site crossover, and the strength of association is presented as odds ratio (OR, 95% confidence interval [CI]). Proportion of access site crossover was evaluated over the study period by quartiles of time in days (total cohort: quartile 1, day 1 to 202; quartile 2, day 203 to 434; quartile 3, day 435 to 702; quartile 4, day 703 to 1,025). Statistical analysis was conducted using the IBM Statistical Package for Social Sciences software, version 20 (IBM Corporation, Armonk, New York) and SAS, version 9.3 (SAS Institute Inc., Cary, North Carolina). Statistical significance was tested using a 2-sided α level of 0.05.
Results
Of the 1,600 patients who met inclusion and exclusion criteria, 166 patients (10.4%) experienced access site crossover. Baseline characteristics of the group that did versus those that did not undergo access site crossover are listed in Table 1 . The proportion of female patients >75 years did not differ between the crossover versus no crossover groups (7.2% vs 5.6%, p = 0.383).
| Variable | Access Site Crossover | p-value | |
|---|---|---|---|
| Yes (n=166) | No (n=1434) | ||
| Age (years) | 61.6 ± 12.4 | 59.7 ± 11.8 | 0.085 |
| Age >75 years | 31 (18.7%) | 177 (12.3%) | 0.028 |
| Women | 57 (34.3%) | 515 (35.9%) | 0.733 |
| White | 48 (28.9%) | 421 (29.4%) | 0.658 |
| Black | 52 (31.3%) | 489 (34.1%) | |
| Hispanic | 30 (18.1%) | 219 (15.3%) | |
| Asian | 24 (14.5%) | 230 (16.0%) | |
| Other | 12 (7.2%) | 75 (5.2%) | |
| Height (cm) | 167.0 ± 11.3 | 167.5 ± 10.4 | 0.784 |
| Weight (kg) | 80.6 ± 17.4 | 83.1 ± 22.0 | 0.416 |
| Body mass index (kg/m 2 ) | 28.9 ± 5.9 | 29.6 ± 7.4 | 0.589 |
| Hypertension ∗ | 130 (78.3%) | 1081 (75.4%) | 0.445 |
| Hyperlipidemia † | 108 (65.1%) | 869 (60.6%) | 0.276 |
| Diabetes mellitus | 67 (40.4%) | 544 (37.9%) | 0.555 |
| Prior percutaneous coronary intervention | 26 (15.7%) | 284 (19.8%) | 0.215 |
| Prior coronary artery bypass graft surgery | 9 (5.4%) | 64 (4.5%) | 0.555 |
| Peripheral artery disease/aortic aneurysm | 9 (5.4%) | 61 (4.3%) | 0.428 |
| Tobacco use | 0.161 | ||
| Never | 120 (72.3%) | 1116 (77.8%) | |
| Prior use | 13 (7.8%) | 70 (4.9%) | |
| Current use | 33 (19.9%) | 248 (17.3%) | |
∗ Hypertension was defined as a clinical diagnosis of hypertension (blood pressure >140/90 or on antihypertensive medications).
† Hyperlipidemia was defined as a clinical diagnosis of hyperlipidemia or on lipid-lowering agents.
Procedural characteristics of access site crossover versus no crossover groups are listed in Table 2 . Timing of access site crossover in the total cohort and by operator experience is listed in Table 3 . Of the patients who did undergo access site crossover, 23 (13.9%) were primarily because of radial artery spasm.
| Variable | Access Site Crossover | p-value | |
|---|---|---|---|
| Yes (n=166) | No (n=1434) | ||
| Right-sided access | 160 (96.4%) | 1358 (94.7%) | 0.457 |
| Attending operators with ≤5 years of transradial experience | 137 (82.5%) | 900 (62.8%) | <0.001 |
| Mean number of catheters used to cannulate right coronary artery | 1.57 ± 0.76 | 1.12 ± 0.40 | <0.001 |
| Mean number of catheters used to cannulate left coronary artery | 1.85 ± 1.24 | 1.16 ± 0.52 | <0.001 |
| Mean number of guide catheters used (PCI patients) | 1.40 ± 0.67 | 1.19 ± 0.52 | 0.003 |
| Contrast used during diagnostic angiography (mL) | 96 ± 48 | 73 ± 54 | <0.001 |
| Contrast used during entire procedure (PCI patients) | 245 ± 112 | 195 ± 81 | 0.001 |
| Variable | Total (n=166) | Operators with ≤5 years TRA experience (n=137) | Operators with >5 years TRA experience (n=29) | p-value |
|---|---|---|---|---|
| Inability to cannulate the radial artery | 47 (28.3%) | 32 (23.4%) | 15 (51.7%) | 0.003 |
| Inability to cannulate any coronary artery (pre-angiography) | 51 (30.7%) | 47 (34.3%) | 4 (13.8%) | 0.021 |
| Inability to complete diagnostic angiography (mid-angiography) | 54 (32.5%) | 46 (33.6%) | 8 (27.6%) | 0.347 |
| For PCI procedure (between diagnostic angiography and percutaneous coronary intervention) | 9 (5.4%) | 8 (5.8%) | 1 (3.4%) | 0.512 |
| After an initial attempt to cannulate the coronary artery with a guide catheter (during percutaneous coronary intervention) | 5 (3.0%) | 4 (2.9%) | 1 (3.4%) | 0.622 |
Univariate predictors of access site crossover included age >75 years (OR 1.63, 95% CI 1.07 to 2.48, p = 0.0228) and operator transradial experience ≤5 years (OR 2.80, 95% CI 1.85 to 4.24, p <0.0001). Although the rate of access site crossover in operators with >5 years of transradial experience (n = 563) was 5.2%, the rate of access site crossover in operators with ≤5 years of transradial experience (n = 1,037) was 13.2%. The proportions of procedures that underwent access site crossover over time, stratified by operator transradial experience, are shown in Figure 1 .
