Introduction
Radial access is the preferred puncture site for ST-segment elevation acute myocardial infarction (STEMI) treatment, due to lower complication rates such as bleeding, shorter hospitalization days, and similar success rates compared to femoral approach1-4. On the other hand, femoral access has several advantages including lower radiation exposure, more familiarity with this technique, more availability of devices, and higher procedural success rates, which have made this access the gold standard for many decades5,6. Nevertheless, controversies still exist regarding which vascular approach is better in diagnostic angiographic studies and non-ST-segment elevation myocardial infarction (NSTEMI)1. Therefore, the aim of this study was to compare the differences in intervention time, fluoroscopy time, frequency, temporal trends, and risk factors for complications, in both radial and femoral access in a third-level hospital in Mexico City.
Materials and methods
This was a cross-sectional study made in the Interventional Cardiology Service from the Hospital Central Sur de Alta Especialidad PEMEX Picacho in Mexico City during October 2009-January 2011. We included 493 patients in which coronary procedures were performed for stable coronary artery disease and acute coronary syndromes (both diagnostic and therapeutic). Before the interventional procedure, Allens maneuver was performed in each patient to examine ulnar collateral flow. The choice of radial or femoral approach was left to the interventional cardiologist discretion. For radial approach with the Seldinger technique, puncture site was identified by palpation of the radial region (2-3 cm proximal to the radial styloid process) and local anesthesia (2% lidocaine, 10-15 ml) was applied to subcutaneous tissue. Puncture was done at 30-40° and the needle was directed toward the flow direction. After obtaining flow, a guidewire was advanced and the needle was withdrawn. A dilator was later advanced through the guidewire and saline solution with heparin was injected. Finally, an introducer sheath was placed for the passage of guide wires or stents, as needed by each patient. For femoral approach with Seldinger technique, we identified the puncture site 2 cm below the inguinal ligament at the intersection between the anterosuperior iliac spine and the pubic symphysis. The same steps as for radial approach were later repeated. We recorded demographic variables such as age, weight, height, and hospitalization days from the medical history, as well as comorbidities such as hypertension, diabetes, chronic stable angina, previous myocardial infarction, hyperuricemia, dyslipidemia, and smoking. Angiographic procedural data such as type of vascular approach (radial or femoral), complications, intervention time, fluoroscopy time, amount of contrast media, dose-product area, and location of artery disease were taken from the interventional cardiology files.
Statistical analysis
Data were analyzed with STATA/IC v13 (StataCorp, College Station, Texas). For the descriptive analysis, binary variables were described as frequencies and proportions, and they were analyzed with Pearsons independence test (χ2) or Fishers exact test, according to the number of individuals per case in the 2 by 2 table. Quantitative variables were analyzed first with Shapiro-Wilks normality test, and according to this, they were described as parametric (mean, standard deviation, and minimum-maximum) or non-parametric (median, interquartile range, and minimum-maximum). Bivariate analysis was done with Students t-test for parametric variables, and with Mann-Whitneys U-test for non-parametric variables. We constructed a logistic regression model for determining the risk factors for complications. p menor 0.05 was considered statistically significant.
Results
We included 493 patients who underwent a percutaneous coronary intervention of diagnostic angiography, of which 70.18% were men and 29.82% were women, with a median age of 63 years, median height of 166 cm, median weight of 77 kg, and median of 5 hospitalization days (Table 1).
Invariables | Overall (n = 491), n (%) | Radial approach (n = 159), n (%) | Femoral approach (n = 334), n (%) | p | |||
---|---|---|---|---|---|---|---|
Men | 346 (70.18) | 131 (82.39) | 215 (64.37) | 0.00 | |||
Women | 147 (29.82) | 28 (17.61) | 119 (35.63) | ||||
Hyperuricemia | 196 (39.75) | 58 (36.71) | 138 (41.82) | 0.28 | |||
CKD | 87 (17.64) | 36 (22.78) | 51 (15.45) | 0.04 | |||
Previous MI | 251 (50.91) | 79 (50) | 172 (52.12) | 0.66 | |||
Dyslipidemia | 245 (49.69) | 72 (45.57) | 173 (52.42) | 0.15 | |||
Smoking | 163 (33.06) | 59 (37.34) | 104 (31.52) | 0.20 | |||
Hypertension | 332 (67.30) | 107 (67.72) | 225 (68.18) | 0.91 | |||
Diabetes mellitus | 237 (48.07) | 74 (46.84) | 163 (49.39) | 0.59 | |||
Stable angina | 224 (45.90) | 71 (44.94) | 153 (46.36) | 0.76 | |||
ACS | 89 (18.24) | 63 (19.09) | 28 (20.14) | 0.48 | |||
Variables | n | Median (IQR) (minimum-maximum) | n | Median (IQR) (minimum-maximum) | n | Median (IQR) (minimum-maximum) | p |
Age (years) | 460 | 63 (55-70) (23-88) | 128 | 62 (54-69.5) (42-83) | 332 | 63 (56-70) (23-88) | 0.35 |
Height (cm) | 437 | 166 (160-170) (144-190) | 112 | 168 (160-170) (144-185) | 325 | 165 (158-170) (140-190) | 0.03 |
Weight (kg) | 438 | 77 (69-83) (35-125) | 112 | 80 (74-85) (51-125) | 326 | 75 (66-82) (35-125) | 0.03 |
Hospitalization days | 488 | 5 (4-8) (2-92) | 157 | 4 (3-8) (2-68) | 331 | 5 (4-8) (2-92) | 0.03 |
CKD: chronic kidney disease; MI: myocardial infarction; ACS: acute coronary syndrome; IQR: interquartile range.
Regarding comorbidities, 50.91% (n = 251) had previous myocardial infarction, 67.3% (n = 332) hypertension, 48.07% (n = 237) diabetes, 63.48% chronic stable angina, 49.69% dyslipidemia (n = 245), 39.7% (n = 196) hyperuricemia, 17.64% (n = 87) chronic kidney disease, and 33.06% (n = 163) were smokers. The subgroup analysis showed significant differences between radial and femoral approach in sex, chronic kidney disease, hospitalization days, height, and weight (Table 1).
From the whole population, 30.08% did not have significant coronary disease or underwent diagnostic angiography for other reasons (i.e. congenital heart disease or valvulopathy); on the other hand, 36.17% had three-vessel disease. The most frequently diseased coronary artery was the left anterior descending, which was affected in 51.5% and 48.2% in the radial and femoral approach, respectively (Table 2).
Variables | Radial approach (n = 159) | Femoral approach (n = 334) | p |
---|---|---|---|
Normal coronary arteries | 28 (20.14) | 71 (23.28) | 0.46 |
1-vessel disease | 33 (23.74) | 52 (17.05) | 0.09 |
2-vessel disease | 26 (18.78) | 55 (18.03) | 0.86 |
3-vessel disease | 52 (37.41) | 126 (41.31) | 0.43 |
Left anterior descending | 82 (51.57) | 161 (48.20) | 0.48 |
Circumflex artery | 68 (42.77) | 130 (38.92) | 0.41 |
Right coronary artery | 68 (42.77) | 150 (44.91) | 0.65 |
Left main coronary artery | 150 (44.91) | 44 (27.67) | 0.29 |
Intermediate artery | 31 (19.50) | 60 (17.96) | 0.68 |
Diagonal arteries | 43 (27.04) | 96 (28.74) | 0.69 |
Marginal arteries | 38 (23.90) | 78 (23.35) | 0.89 |
The femoral vascular access was more commonly used in this study 334 interventions. Complications appeared 18 times, which represented 3.65% of the total population. The most frequent complication was vasospasm (n = 9), which occurred mainly in the radial approach, followed by hematoma (n = 7): 3 in the radial and 4 in the femoral access. Crossover from radial to femoral approach happened in 9 times and was always associated to complications in the original puncture site.
Complications were significantly higher in the radial approach 6.92% versus 2.1% in the femoral approach. Intervention time between both techniques was similar; nevertheless, fluoroscopy showed significant differences because the radial approach had 3 min more than the femoral. Finally, dose-area product was significantly higher in the femoral approach (Table 3).
Variables | Radial approach (n = 159) | Femoral approach (n = 334) | p | ||
---|---|---|---|---|---|
Complications | 11 (6.92) | 7 (2.10) | 0.00 | ||
Hematomas | 3 (1.89) | 4 (1.20) | 0.54 | ||
Vasospasm | 8 (5.03) | 1 (0.30) | 0.00 | ||
Ventricular fibrillation | 0 (0) | 2 (0.60) | 0.32 | ||
Death | 0 (0) | 0 (0) | 1 | ||
Crossover | 9 (5.66) | 0 (0) | 0.00 | ||
Variables | n | Median (IQR) (minimum-maximum) | n | Median (IQR) (minimum-maximum) | p |
Intervention time (hh: mm: ss) | 143 | 1:20:00 | 315 | 1:20:00 | 0.85 |
Fluoroscopy time (hh: mm: ss) | 143 | 00:12:00 | 315 | 00:09:01 | 0.05 |
Contrast media (ml) | 77 | 140 (120-250) (80-700) | 124 | 122.5 (100-180) (50-800) | 0.00 |
Dose-product area (μG m2) | 143 | 2282 (1128.7-7290) (316.4-84,400) | 315 | 2800 (1034-5600) (79-24,120) | 0.04 |
IQR: interquartile range.
In the logistic regression model, we found that intervention time longer than 60 min was associated with an increase in the overall occurrence of complications. Furthermore, we found certain tendencies toward risk increase for one-vessel disease, diagonal arteries disease, circumflex artery disease contrast media > 150 ml, dyslipidemia, left main artery disease, fluoroscopy > 10 min, female sex, and smoking, but without statistical significance (Table 4).
Variables | OR | SE | p | 95% CI |
---|---|---|---|---|
1-vessel disease | 6.19 | 7.69 | 0.14 | 0.54-70.76 |
Total intervention time > 60 min | 5.49 | 4.59 | 0.04 | 1.06-28.32 |
Diagonal arteries | 3.93 | 3.25 | 0.09 | 0.77-19.89 |
Circumflex artery disease | 3.80 | 3.22 | 0.11 | 0.72-19.98 |
Dyslipidemia | 2.75 | 1.93 | 0.15 | 0.69-10.92 |
Contrast media > 150 ml | 1.62 | 1.41 | 0.57 | 0.29-8.94 |
Left main artery disease | 1.54 | 1.49 | 0.65 | 0.23-10.28 |
Fluoroscopy time > 10 min | 1.52 | 1.32 | 0.62 | 0.27-8.41 |
Female sex | 1.28 | 0.91 | 0.72 | 0.31-5.22 |
Smoking | 1.07 | 0.73 | 0.92 | 0.27-4.11 |
2-vessel disease | 1.00 | 1.54 | 0.99 | 0.04-20.60 |
3-vessel disease | 0.94 | 1.63 | 0.97 | 0.03-27.79 |
Right coronary artery disease | 0.94 | 0.70 | 0.94 | 0.21-4.08 |
Previous myocardial infarction | 0.91 | 0.56 | 0.87 | 0.26-3.10 |
Left anterior descending disease | 0.80 | 0.58 | 0.76 | 0.19-3.30 |
Age less 50 years | 0.75 | 0.71 | 0.76 | 0.11-4.88 |
Intermediate artery disease | 0.52 | 0.65 | 0.60 | 0.04-3.96 |
Diabetes mellitus | 0.47 | 0.33 | 0.28 | 0.12-1.85 |
Marginal arteries | 0.40 | 0.47 | 0.43 | 0.04-3.96 |
Chronic kidney disease | 0.28 | 0.32 | 0.26 | 0.03-2.57 |
Acute coronary syndrome | 0.22 | 0.27 | 0.22 | 0.01-2.46 |
R2= 0.3. OR: odds ratio; SE: standard error; 95% CI: 95% confidence interval.
Temporal trends for the presence of complications showed that 14 events occurred in the first 6 months, and the 3rd month had the largest amount of events (n = 5). Figure 1 shows the frequency of complications per month and a decreasing tendency of these events toward the end of the study.
Discussion
To the best of our knowledge, this is the first study to compare outcomes by either radial or femoral approach in a Mexican population. Campeau described radial access for the 1st time in 1989, and its use has been increasing since then1-6. The main anatomical advantages offered for coronary interventions are the absence of large-size nerves or veins near the puncture site and double circulation to the rest of the hand which prevents ischemia and movement complications.7 Furthermore, it has become the most used technique due to high success and low complication rates, along with bleeding reduction in the puncture site. Nowadays, due to a more extended use, the radial access is not only applied in coronary interventions, but in peripheral interventions such as carotid artery, femoral, superficial, mesenteric, and renal artery disease3.
Our study found that the complication rate was higher among radial access patients compared to femoral approach; nevertheless, we have to clearly state that most of these events (n = 14) occurred during the first 6 months of the implementation of this technique. This phenomenon can be explained by the learning curve period, which is inherent to every recently acquired technique, in which the center has to perform certain amount of procedures to gain experience to accomplish a high success rate and low complication frequency as explained in other studies8,9, On the other hand, it is important to recall that the study was made in an Interventional Cardiology Department that formerly was adapted to perform procedures using the femoral approach, and the operators had to adapt to the new radial-designed materials such as catheters, guide wires, and introducer sheaths.
As recently shown by the most recently published guidelines on ST-elevation myocardial infarction, radial approach has become the default access for coronary interventions in this context, due to its higher success, lower bleeding, and lower mortality rates1. It is important to recall that our study included acute coronary syndromes as well as diagnostic angiographies and procedures for chronic stable angina. Furthermore, it is important to mention the great amount of multivascular disease (i.e., disease of more than one vessel) to state a more complex coronary anatomy in this population, which increases the intervention and fluoroscopy times, as well as the overall complication rates including bleeding and vasospasm, and prolongs the intervention time as well as the fluoroscopy time.
Vascular complications are the most prevalent and are mainly provoked by anatomical variants in the puncture site10-12. After multiple punctures, radial artery tends to thicken its intimal media layer13; this generates higher complication and lower success rates. Unfortunately, we were not able to record which patients underwent more than one puncture or have had a previous coronary intervention by the radial approach.
Vasospasm was the most commonly found complication in both types of approach. Meanwhile, crossover was done 9 times due to complication in the radial access. Radial access failure has been reported in up to 12.5%, and the most common factors are the lack of guide wire support, difficulty to cannulate the vessel and unfavorable anatomy14.
In our analysis, women made up to almost 30% of the population, and the gender did not explain the presence of complications. Other studies have demonstrated that women are more prone to require more than one radial puncture and that they have a higher risk for bleeding compared to men, which can be explained due to a less prominent diameter of the radial artery15,16.
Radiation exposure is one of the main concerns for both the interventional cardiologist and the patient17. Radial access is associated to a higher radiation exposure, measured in µG m2, in diagnostic and angioplasty procedures18,19; however, in our study, radiation exposure was higher in the femoral access, suggesting that radial access can be both safe and effective. Regarding interventional and fluoroscopy time, the previous studies showed longer exposure times in radial access20-23. This fact can be explained because many of these studies were performed before 2010, a time in which a lower frequency of radial procedures was performed. On the other hand, our study showed that there were differences in the fluoroscopy time, with lower times in the femoral approach; meanwhile, the logistic regression model that intervention times longer than 60 min was associated to more complications.
One of the most acclaimed advantages for the radial access is the reduction in hospitalization time due to a lower need of resting time and a faster time to the reintegration to the daily activities24. In our study, we were able to demonstrate a 1-day reduction in hospitalization, which can lower the costs of health systems and help to a faster reintegration to the patients life.
Finally, we have to make emphasis that both approaches are safe and that randomized controlled trials show that in high-volume centers, in which a greater amount of procedures are made, radial access procedures have shown better results25.
The limitations of our study were the lack of randomization that could have made the groups more homogeneous, the high number of interventional cardiologist operators and being a single-center study.
Conclusions
This study showed that femoral approach had lower complication rates than the radial approach. Furthermore, the complications occurred more frequently in the first 6 months and the main predictor for complications was the intervention time > 60 min. More studies are needed to identify earlier and prevent the potential complications associated to interventional procedures.