Percutaneous coronary intervention for unprotected left main coronary artery : radial vs . femoral access

Background: Percutaneous intervention for obstructive lesions of the unprotected left main coronary artery is complex. Choosing the vascular access can be decisive to a successful procedure. The objective of the present study was to describe the characteristics and hospital outcomes of patients with unprotected left main coronary artery disease and compare the radial and femoral access approaches. Methods: Clinical and angiographic data, access routes and hospital outcomes of patients that have undergone percutaneous coronary intervention of the unprotected left main coronary artery were collected from the Central Nacional de Intervenções Cardiovasculares (CENIC) between June 2006 and March 2016. Results: A total of 734 patients were included. Mean age was 66±12 years, 62% were male and 22% were diabetic. Acute coronary syndrome occurred in 47%. Cardiogenic shock was observed in 44% of patients with acute myocardial infarction. The procedure success rate was 90% and the hospital mortality rate was 8.5%. Femoral access was the most frequent approach. Regardless of the vascular approach used, there was no difference in the rate of major adverse cardiovascular events during hospital stay. Conclusions: High clinical complexity was observed in CENIC patients presenting with unprotected left main coronary artery lesions treated by percutaneous coronary intervention. Femoral access was the most frequent, especially for more complex cases. There was no difference between the procedure success rate and adverse event rate when comparing both vascular approaches.


IntroductIon
Obstructive lesions of the unprotected left main coronary artery (LMCA), especially those involving its distal portion, are complex percutaneous procedures,

Journal of Transcatheter Interventions
2 requiring the use of multiple balloons and interventional techniques with two or more stents.The choice of vascular access may determine the success of the procedure in this setting.
The femoral approach allows using larger diameter catheters (7 and 8 Fr), providing greater support and facilitating the treatment of bifurcations, especially when using balloons and large diameter stents simultaneously.Despite presenting a shorter learning curve, the femoral technique is associated with a higher rate of vascular and hemorrhagic complications. 1,2 e radial approach is much safer, reducing the occurrence of complications related to the arterial puncture site, and also reducing mortality in the context of acute coronary syndromes. 3][6][7] The objective of the present study was to describe clinical and angiographic characteristics and hospital outcomes of patients undergoing percutaneous coronary intervention (PCI) of the unprotected LMCA, comparing the radial and femoral access approaches.metHods This is a retrospective study with information collected from data registered in the Central Nacional de Intervenções Cardiovasculares (CENIC; http://www.corehemo.net/), of the Sociedade Brasileira de Hemodinâmica e Cardiologia Intervencionista (SBHCI), created in 1991.The registry has information on PCI procedures gathered on a dedicated database, entered voluntarily by member physicians from various Brazilian institutions.The study was approved by the Hospital Leforte Ethics and Research Committee, as CAEE 90669318.2.0000.5485.
Data for the study were collected from 734 patients presenting unprotected LMCA disease treated with PCI, from June 2006 to March 2016.Patients were divided into two groups, radial and femoral, according to the access site chosen by the physician performing the procedure.Groups were compared in relation to clinical characteristics (age, sex, systemic arterial hypertension, diabetes mellitus, dyslipidemia, previous PCI, previous myocardial infarction, clinical presentation and Killip classification), angiographic data (extent of coronary artery disease, anatomical complexity, pre-procedure Thrombolysis in Myocardial Infarction -TIMI flow, presence of left ventricular dysfunction and presence of collateral circulation), and procedure data (frequency of stent use; percentage of drug-eluting stent use; length, diameter and number of stents per patient; type of intervention performed; use of thromboaspiration; use of glycoprotein IIb/IIIa inhibitors; post-procedure TIMI flow; degree of pre-and post-procedure stenosis; and procedure success rate).Clinical hospital outcomes analyzed were death, emergency coronary artery bypass graft sur-gery (CABG), acute myocardial infarction (MI) and major adverse cardiovascular events (MACE).
The patient was considered as the sample unit for the analysis of clinical variables.Some patients underwent more than one procedure.In these cases, the procedure to be considered was randomly selected.The Chi-square test was used to compare categorical variables.The odds ratio was used whenever necessary.Analysis of variance (ANOVA) was used to compare continuous variables.The Bonferroni correction was used for multiple comparisons.To verify the influence of variables of interest on mortality of patients treated with PCI of unprotected LMCA, simple and multiple logistic regression models (forward) were used.A significance level of 5% (p-value <0.05) was set for all analyses.

results
A total of 734 patients (576 femoral and 158 radial approaches) were included in the study.A total of 739 procedures were performed, with a mean of 1.01 per patient.A total of 769 vessels were treated, 725 (94.3%) of which with stents, totaling up 764 stents.There were significant differences between the groups regarding clinical characteristics (Table 1).The group with femoral access had a higher percentage of smokers and older patients.Regarding clinical status, stable angina and asymptomatic patients prevailed in the group with radial access, while acute MI and non-ST segment elevation acute coronary syndrome (NSTE-ACS) were more prevalent in the femoral group.Patients in the Killip 4 class were more frequent in the femoral approach.
Regarding angiographic characteristics (Table 2), patients in the femoral access group had a higher percentage of calcified, long lesions, with a higher thrombus load and a pre-procedure TIMI coronary flow grade 0/1.There was no significant difference in the extent of coronary disease between groups.It is noteworthy that in 44 patients with femoral access and in 40 with radial access, there was no record of the extent of the coronary disease.
Regarding the characteristics of the procedure (Table 3), patients in the femoral access group had a higher proportion of vessels treated per patient, and also of primary and rescue PCI than those with radial access.The radial approach patients had a higher stent-to-patient ratio, higher use of drug-eluting stents and longer devices than those with femoral access.

4
There was no significant difference in clinical hospital outcomes (Table 4).Occurrence of acute MI, need for revascularization, death and MACE were similar between the groups.In the simple logistic regression model (Table 5), some variables were associated with death: clinical presentation of acute MI, primary PCI, use of glycoprotein IIb/IIIa inhibitor, presence of left ventricular dysfunction, Killip 4 class, and extension of coronary disease.Arterial hypertension, dyslipidemia and history of previous PCI were predictors of death.Vascular access (radial or femoral) was not associated with the mortality outcome.Variables with incomplete data (Killip, ventricular dysfunction and collateral circulation) were excluded from the multivariate logistic regression analysis (Table 6).The independent variables most associated with death were extension of coronary disease and type of procedure performed (primary PCI).

dIscussIon
There are few studies comparing radial and femoral approaches for PCI of unprotected LMCA.Our work compared both vascular access sites retrospectively, using the CENIC database, which reflects the national scenario.Femoral access was the most common access site for the treatment of unprotected LMCA in the Brazilian registry.Patients with more severe clinical characteristics (elderly, Killip 4 functional class and with acute coronary artery disease) were predominantly treated by the femoral approach.The radial approach had a higher proportion of young patients, asymptomatic patients or those with stable an gina, and functional class 1/2.
Although the radial approach is safer, with less hemorrhagic and vascular complications, it has a longer learning curve, being reserved initially for less complex cases and stable patients. 6,7Stronger scientific evidence showing reduction of bleeding and mortality, and recommendations in guidelines encouraging the use of the radial approach are relatively recent. 1,2These factors may have influenced this approach to be less used for treating unprotected LMCA lesions in the CENIC registry, during the investigation period.In addition, although the technical limitations imposed by the radial approach (smaller caliber, less support) can be overcome with the use of techniques (buddy wire, buddy balloon and anchorage) and dedicated devices (guide catheters with greater support and mother and child catheter), these devices are often not widely available in Brazilian cath labs.
The angiographic characteristics of the lesions reflected the clinical profile of each vascular access group.Long lesions

5
with a marked degree of calcification were more frequently treated by the femoral approach; lesions with a high thrombotic load and pre-procedure coronary TIMI flow grade 0/1, as well as interventions, such as primary and rescue PCI, prevailed in the femoral group, and may be explained by the higher percentage of ACS and acute MI in this group.
Our analysis of the procedures showed that a small number of stents per procedure was used, suggesting low use of techniques with multiple devices.The proportion of drug-eluting stents was higher when using the radial technique, which may be explained by the higher percentage of elective cases treated in this access route.
][10][11] The multivariate logistic regression model showed that coronary artery disease and primary PCI were significantly associated with death, corroborating findings in the literature. 12ur study has important limitations.It is a retrospective, non-randomized analysis with operator selection bias.The hemorrhagic and vascular complications, amount of contrast used, dose of radiation, length of hospital stay and crossover rate between the vascular approaches were not accounted for; neither were the frequency of multiple stent techniques nor success in the final kissing balloon in procedures with two or more stents.conclusIons Patients on the CENIC registry with unprotected left main coronary artery lesions treated with percutaneous coronary intervention were of high clinical complexity.Femoral access was the most frequent, especially in more complex cases.There was no difference in the success rate of the procedures or in major adverse cardiovascular events when the radial and femoral access routes were compared.

Table 2 .
Angiographic characteristics Results expressed n (%).*The extension of the coronary disease was not reported in the registry of some patients.LMCA: left main coronary artery; TIMI: Thrombolysis in Myocardial Infarction.

Table 4 .
Clinical outcomes during hospital admission MI: myocardial infarction; CABG: coronary artery bypass surgery; MACE: major adverse cardiovascular events.

Table 6 .
Determinants of mortality by multiple logistic regression