Comparison of the Radial and Femoral Approaches in Left Main PCI: A Retrospective Study

ABSTRACT: Transradial percutaneous coronary intervention (PCI) is a safe and effective method of percutaneous revascularization. However, there are no data on the efficacy of the transradial approach in left main (LM) PCI. We studied 80 patients (pts) who underwent LM PCI between February 1994 and January 2002, and compared the radial (27 pts) and femoral (53 pts) approaches. Patients were considered free of restenosis if they were free of angina and had a negative treadmill or nuclear imaging study 6 months post-PCI. Mean follow-up time was 27.4 ± 23.0 months. Reason for PCI (stable angina, unstable angina, acute myocardial infarction) and lesion location (ostial, mid, distal) were similar in both groups (p > 0.05), whereas mean ejection fraction was higher in the radial group (56.5 ± 11.1% versus 49.2 ± 14.7%, respectively; p p p > 0.05). However, major vascular complications occurred only in the femoral group (5.7%). Radial LM PCI is as fast and successful as the femoral approach and results in fewer vascular complications.

Key words: left main, PCI, radial

Coronary artery bypass grafting (CABG) is the standard treatment for left main (LM) coronary artery disease.^1,2 Balloon angioplasty of the LM has been associated with poor results.^3–7 However, new percutaneous technologies,^8–10 especially stenting,^11–23 have shown favorable results for unprotected and protected LM percutaneous coronary intervention (PCI). Transradial angioplasty with stenting is a safe and effective percutaneous revascularization method.^24–26 However, there are no available data for the use of the radial approach for LM PCI. This retrospective study evaluated 80 patients who underwent LM PCI through the radial or femoral approach in our hospital and compared efficacy and safety. Methods Study population. We retrospectively reviewed 80 patients who underwent PCI in our hospital between February 1994 and January 2002 due to significant LM disease. Significant LM disease was defined as clinical symptoms or objective evidence of myocardial ischemia and LM diameter stenosis of > 50% on angiography. Unprotected LM was defined as significant LM stenosis without the presence of a patent bypass graft to the left coronary artery [left anterior descending (LAD) or left circumflex (LCX)]. Angioplasty was performed using either the radial or femoral approach. Baseline clinical demographics, reason for proceeding to PCI instead of CABG, procedural data and outcome, and in-hospital complications were confirmed by individual hospital chart review. Clinical follow-up was obtained by telephone interviews of patients and by contacting the referring doctor and cardiologist. The major adverse cardiac event (MACE) rate included cardiac death, acute myocardial infarction (AMI) and target lesion revascularization (either CABG or repeat PCI) for a restenotic LM lesion > 50%. Patients were considered free of restenosis only if they were free of symptoms and had a negative non-invasive evaluation (treadmill or nuclear imaging study). If a patient underwent a repeat procedure (PCI or CABG) or experienced an AMI, the hospital files were reviewed. All deaths were reviewed using hospital files and records from the referring physician and cardiologist. The diagnosis of myocardial infarction (MI) in the absence of new pathologic Q-waves on the electrocardiogram was based on creatine phosphokinase-MB elevation > 5 times the normal value. Cardiac death included sudden death and death due to MI or congestive heart failure. Vascular complications due to intra-aortic balloon pump insertion were not included in the vascular complications. Radial approach. In our hospital, the radial approach is the preferred method for PCI. Sixty percent of all procedures (angiograms/PCI) are performed transradially. All patients undergoing radial PCI in our hospital are checked for adequate collateral circulation (Allen test). Radial artery cannulation is accomplished using a Cook needle with a small 21-gauge cannula and an 0.018´´ guidewire. Six or 7 French sheaths are used and a vasodilating cocktail containing 200 µg of nitroglycerin and 1 mg of verapamil is usually administered. After successful PCI, the sheath is immediately removed and a pressure clamp is applied. The clamp is removed within 2–3 hours if hemostasis has been achieved. Femoral approach. Femoral artery cannulation in our hospital is performed using standard techniques, Judkins catheters and 6, 7 or 8 French sheaths. The sheaths are removed 3–4 hours after the end of the procedure regardless of the use of IIb/IIIa inhibitors. Hemostasis at the puncture site is achieved by manual compression, which is usually applied for 20–30 minutes. After achieving hemostasis, a sandbag is applied for another 3–4 hours. Angioplasty. PCI was performed under local anesthesia and sedation. The guiding catheters used included Judkins right or left, XB, Kimny and Amplatz. 0.014´´ coronary guidewires and rapid-exchange balloons were used. Balloon-mounted stents or hand-crimped stents were used. The type of stents used included Minicrown, GR2 Cook, JJ PS 153 or 104, Niroyal, Duet, Tristar, AVE S670 and S7, Velocity, Multi-Link Tetra or Penta, Express and Jostent. The balloon inflations were brief in order to avoid prolonged ischemia and ischemia-related complications. The use of an intra-aortic balloon pump (IABP) was at operator discretion. Successful PCI was defined as a residual diameter stenosis Anticoagulation and antiplatelets. All patients were pretreated with aspirin 325 mg. Intravenous heparin 5,000–10,000 IU in total were given with a target activated clotting time (ACT) of 250–300 seconds. Target ACT for patients receiving abciximab (ReoPro®; Eli Lilly & Company, Indianapolis, Indiana) was 200–250 seconds. When abciximab was administered, a bolus of the drug (0.25 mg/kg body weight) was given, followed by an infusion (0.125 µg/kg/minute) for 12 hours. Clopidogrel or ticlopidine were administered either prior to or at the beginning of the procedure and continued for 4 weeks following stent implantation. Angiographic analysis. Quantitative and qualitative angiographic analyses were obtained for all patients using Quanticore (Siemens, Germany). Percent diameter stenosis (%DS), minimal lumen diameter (MLD) and reference diameter were measured before predilation, after stenting or non-stent procedures, or after final adjunct balloon inflation using a matched view showing the MLD. Angiographic restenosis was defined as diameter stenosis > 50% at follow-up. Statistical analysis. Continuous variables are presented as means and standard deviations. Categorical data are presented as percent frequencies. SPSS 10 for Windows was used for statistical analyses. Differences between group means were tested by the Student’s t-test. Categorical variables were compared using Chi-square analysis. Means in more than 2 groups were compared using A NOVA. A p-value Study population. Eighty patients (60 men; 75.0%) underwent LM PCI in our hospital between February 1994 and January 2002. Forty-four of them (55.5%) had unprotected and 36 (44.5%) had protected LM PCI. Seven patients (8.8%) underwent LM PCI because they presented to the hospital with an AMI (5 patients with unprotected LM, 11.4% and 2 patients with protected LM, 5.6%; p > 0.05). Ten patients (12.5%) had a major dissection in the LM while they were undergoing coronary angiography or PCI, and the operator proceeded to LM PCI (6 patients with unprotected LM, 13.6% and 4 patients protected LM, 11.1%; p > 0.05). The rest of the patients (33; 75.0%) with unprotected LM disease underwent LM PCI instead of CABG because they were considered at very high surgical risk, had limited life expectancy because of other coexisting disease, or were hemodynamically unstable. Mean time of follow-up was 27.4 ± 23.0 months. The baseline clinical characteristics of the patients are presented in Table 1. There were no significant differences in overall baseline clinical characteristics between the radial and femoral groups, except that patients having the femoral approach had significantly more often a history of previous PCI (52.8% versus 29.6%) and significantly lower EF (49.2 ± 14.7% versus 56.5 ± 11.1%; p Procedural data and outcome. Procedural data are presented in Table 2. Twenty-seven patients (33.8%) had the radial approach. The percentage of unprotected LM PCI (16 radial patients, 59.3% versus 28 femoral patients, 63.6%), location of LM stenosis, percentage of occluded RCA, number of significantly diseased coronary vessels, mean reference diameter, mean lesion length and percent stenosis were all similar in the 2 groups. Furthermore, fluoroscopy time, amount of contrast used, administration of IIb/IIIa inhibitors, rate of predilatation, stenting and post-stent dilatation, and the percentage of patients who needed basic cardiac life support or administration of cardiac inotropes during the procedure were also similar. In the femoral group, sheath size (7 or 8 Fr; 44.4% versus 77.3%) and the amount of heparin used (9,192 ± 3,645 versus 11,468 ± 5,083 IU) were significantly larger (p p > 0.05). During hospitalization, two patients (3.8%) died in the femoral group. Both patients presented with cardiogenic shock due to AMI and underwent successful LM PCI, but died because of pump failure. Two patients (7.4%) in the radial group and 1 patient (1.9%) in the femoral group experienced a non-Q wave MI. Major vascular access complications occurred only in the femoral group (5.7%). One patient required transfusion because of large groin hematoma, one had surgical repair of his femoral artery, and 1 had a large hematoma, which prolonged hospital stay. MACE at 6 months were also similar (14.8% for radial group versus 25.5% for femoral group; p > 0.05). In the radial group, four patients (14.8%) had target vessel revascularization (all repeat PCI), and in the femoral group 1 patient (2.0%) had cardiac death and 12 patients (23.5%) underwent target vessel revascularization [7 patients (13.7%) with repeat PCI and 5 (9.8%) with CABG]. There were 3 non-cardiac deaths in the radial group (11.1%) and 4 (7.8%) in the femoral group (4 patients died from cancer, two from chronic obstructive pulmonary disease and 1 from chronic renal failure). Discussion Significant LM coronary artery disease is of critical importance.^27–29 Medical therapy of patients with LM disease is associated with a poor prognosis,^30,31 and CABG is the standard treatment.^1,2 Studies of balloon angioplasty of LM reported varying degrees of procedural success, but uniformly poor long-term results.^3–7 However, advances in percutaneous technology and especially stenting made several centers reconsider the role of percutaneous treatment for LM disease, and favorable results for LM PCI were reported.^8–23 Our study is the first to report the outcomes of the transradial approach use in LM PCI. Procedure and 6-month outcomes. Transradial PCI for patients with stable angina, acute coronary syndromes and AMI has been demonstrated to be as successful as the transfemoral approach.^24–26,32 Our study suggests that transradial LM PCI performed by experienced operators appears to be as effective as the transfemoral, since procedural success, in-hospital MACE and 6-month MACE were all similar. Our study is retrospective, so an important issue is whether patients in the 2 groups were similar. The overall baseline clinical, and angiographic characteristics were almost all similar in the 2 groups. Patients in the radial group had significantly higher EF; however, the clinical importance of this difference may not be relevant, because in the largest study of unprotected LM PCI,18 mortality was significantly greater only when the EF was less than 40% (mortality, 1.7% with EF > 40% and 31.8% with EF p > 0.05). Technical considerations. It is well known that distal LM PCI is more demanding, because the origin of the LAD and LCX are commonly involved, requiring the use of kissing techniques or bifurcation stenting. In our study, the percentage of distal LM PCI was similar in the radial and femoral groups (70.4% versus 69.8%, respectively; p > 0.05). This is interesting, because 6 French sheaths were used in more than half of the radial procedures (55.6%). This shows that the availability of smaller diameter equipment has made the treatment of bifurcation lesions feasible through smaller diameter sheaths. Procedure duration. LM PCI is definitely the most demanding percutaneous intervention. Procedure success should be achieved without any delay in interventional maneuvers that can prolong ischemia. In our study, procedure duration was similar with the radial and femoral approaches. This finding is consistent with the results of previous studies, which found similar procedure times for radial and femoral PCI.^25,32,33Vascular complications. An important issue of all percutaneous interventions is the risk of access site complications. A consistent finding in all studies comparing the transradial and femoral approaches for PCI is that access site complications are significantly lower in the radial procedures.^25,26,32,33 In our study, major vascular complications occurred only in the femoral group (5.7%), and this rate is similar to these studies. However, it is important to mention that larger amounts of heparin and larger size sheaths were used in the femoral group, which can partially explain the difference in the rate of vascular complications. Study limitations. The major limitation of our study is that it is a retrospective study and not randomized. One could argue that access site might have been decided based on patient clinical status. Randomized studies can provide further information. Another important limitation of our study is the lack of angiographic follow-up. However, all patients were considered free of restenosis if they were free of symptoms and had a negative non-invasive evaluation (treadmill or nuclear imaging study). Another limitation is the relatively small number of patients. However, the utilization of LM PCI is very small, since CABG is the usual treatment for these patients. Larger multicenter studies will be useful for further information. Clinical implications. Our study suggests that the transradial approach used by experienced operators can be an effective approach in LM PCI for patients not in cardiogenic shock, and its use probably leads to fewer vascular complications. Acknowledgments. The author would like to thank P. Forman, S. McCauley and K. Theman for their help in the collection of data, and data analysis.