Summary
Background
Literature suggests that radial access is associated with higher radiation doses than femoral access.
Aims
To compare patient radiation exposure during coronary angiography (CA) and percutaneous coronary intervention (PCI) with radial versus femoral access.
Methods
RAY’ACT is a nationwide, multicentre, French survey evaluating patient radiation in interventional cardiology. Variables of patient exposure from 21,675 CAs and 17,109 PCIs performed at 44 centres during 2010 were analysed retrospectively.
Results
Radial access was used in 71% of CAs and 69% of PCIs. Although median fluoroscopy times were longer for radial versus femoral access (CA, 3.8 vs 3.5 minutes [ P < 0.001]; PCI, 10.4 vs 10.1 minutes [ P = 0.001]), the Kerma-area product (KAP) was lower with radial access (CA, 26.8 vs 28.1 Gy·cm 2 ; PCI, 55.6 vs 59.4 Gy·cm 2 ; both P = 0.001). Differences in KAP remained significant in the multivariable analysis ( P < 0.01), and in a propensity score-matched analysis ( P = 0.01). A significant interaction was found between KAP and the percentage of procedures with radial access by centre ( P < 0.001). KAP was higher by radial versus femoral access in low-radial-volume centres, and lower in high-radial-volume centres. Radiation protection techniques, such as the use of low frame rates (7.5 frame/s), were used more frequently in high-radial-volume radial centres.
Conclusions
In this multicentre study, radial access was associated with lower radiation doses to patient than femoral access in high-radial-volume centres. Provided that radioprotection methods are implemented, radial access could be associated with lower patient radiation exposure.
Résumé
Contexte
La voie radiale est considérée comme plus irradiante pour les patients que la voie fémorale pour les coronarographies (CA) et les angioplasties coronaires (PCI)
Objectifs
Cette étude a comparé l’exposition des patients lors de CA et PCI réalisées par voie radiale et fémorale.
Méthodes
RAY’ACT est une étude française multicentrique évaluant l’exposition des patients aux rayons x lors des procédures de cardiologie interventionnelle coronaire, qui a analysé rétrospectivement les paramètres d’exposition pour 21 675 CA et 17 109 PCI réalisées dans 44 centres en 2010.
Résultats
La voie radiale a été utilisée dans 71 % des CA et 69 % des PCI. Bien que le temps de scopie médian soit plus long pour la voie radiale versus fémorale (CA, 3,8 vs 3,5 minutes [ p < 0,001] ; PCI, 10,4 vs 10,1 minutes [ p = 0,001]), le produit dose-surface (KAP) était inférieur pour la voie radiale (CA, 26,8 vs 28,1 Gy·cm 2 ; PCI, 55,6 vs 59,4 Gy·cm 2 ; tous p = 0,001). Les différences de KAP restaient significatives en analyse multivariée ( p < 0,01), et par analyse stratifiée sur un score de propension ( p = 0,01). Une interaction significative a été observée entre le KAP et le pourcentage de procédures par voie radiale des centres ( p < 0,001). Le KAP était plus élevé par la voie radiale vs fémorale dans les centres à faible volume pour la voie radiale, et inférieur dans les centres à haut volume pour cette voie. Les techniques de radioprotection, telles que les cadences faibles (7,5 images/s), étaient plus souvent utilisées dans les centres à haut volume pour la voie radiale.
Conclusions
Dans cette étude multicentrique, la voie radiale était associée à une irradiation du patient inférieure à la voie fémorale, globalement et particulièrement dans les centres à haut volume de voie radiale. Sous réserve de la mise en place de méthodes de radioprotection, la voie radiale pourrait être associée à une irradiation du patient réduite.
Background
Since its introduction in 1989, radial arterial access has been used increasingly for coronary angiography (CA) and percutaneous coronary intervention (PCI) in many European countries and, to a lesser extent, in the USA . Transradial PCI has been associated with a lower bleeding risk and fewer access-site complications compared with femoral access . In the setting of acute ST-segment elevation myocardial infarction, the radial route for primary PCI has been associated with improved survival . However, unresolved issues related to the radial access remain, including radiation exposure . Various studies suggest that transradial CA and PCI are associated with higher radiation doses compared with the femoral route, to both patients and staff . Radiation dose can be modulated by the learning curve for transradial interventions , and the procedural volume of the operator and centre . Because reduction of radiation during cardiac procedures is mandatory , overexposure may be a limitation of radial access.
Patient’s Exposure to X-Ray During Coronary Angiography and Percutaneous Transluminal Coronary Intervention (RAY’ACT) is a large, nationwide, multicentre survey aimed at evaluating practices in patient radiation protection in France, a country where the radial route is highly used. In a preliminary analysis, we identified factors associated with the between-centre differences in patient exposure . Raw analyses found that the radial access was associated with lower radiation doses to the patient, and this result contrasted with others in a recent meta-analysis . The purpose of the present study was to further analyse the relationship between radiation dose and arterial access, by comparing radial and femoral routes for variables related to patient exposure, and by testing the roles of the volume of the centre for radial access and the use of techniques for protection from radiation.
Methods
Study design
RAY’ACT-1 is a French, nationwide, investigator-driven, industry-independent, observational, retrospective study, conducted in 44 interventional cardiology centres in France . Patient identities were preserved according to current ethical regulations. The study protocol was approved by national ethics committees and the institutional committee on human research, and subjects provided informed consent.
Data collection
Data from CAs and PCIs performed from 01 January to 31 December 2010 were collected from 44 centres using local software. For each procedure, the following data were collected: patient characteristics (sex, age, body mass index [BMI]); examination details, including arterial access (radial, femoral; the other brachial accesses such as humeral and ulnar were attributed to the radial group); and dosimetry indicators (Kerma-area product [KAP], fluoroscopy time, number of acquisition runs and number of frames) . The radiological equipment comprised 48 cardiovascular X-ray imaging systems (four centres have two catheterisation laboratories, 40 centres have one catheterisation laboratory) from four different manufacturers, installed between 1998 and 2010, 79% of which had a flat panel detector. The choice of the arterial route was at the operator’s discretion. Occupational dosimetry data, registered on separate computer systems with limited access, were not available.
Statistical analysis
Categorical variables are presented as counts and percentages, and were compared using the χ 2 test. Continuous data are presented as medians [interquartile ranges], and were compared using the non-parametric Mann-Whitney U test. Log-normally distributed continuous data, such as KAP and fluoroscopy time, were log-transformed for univariate and adjusted comparisons between radial and femoral routes by analysis of variance, and for linear correlation analyses.
Given the observational design of the study, and to minimize indication bias for arterial access, propensity score analyses were conducted. We estimated the propensity score of having a radial access, by fitting a logistic regression model using age, sex, BMI, emergency procedure and performance of left ventriculography as covariates. We then matched the patients who underwent a procedure via radial access with those who had femoral access, by stratification into subsets based on the quintiles of the estimated propensity score. Patients who could not be matched using these criteria were removed from the analysis. Then, analyses of the association between the arterial access and patient exposure were repeated after matching within each propensity score stratum (13,655 CAs and 8816 PCIs).
To assess the role of centre volume of transradial procedures in the relationship between radial access and radiation dose, a term of interaction was introduced, and was tested in the multivariable linear model. Significance levels were adjusted for multiple comparisons using a conservative Bonferroni’s correction. P -values were two-sided and were considered statistically significant at < 0.01. All statistical analyses were carried out with IBM SPSS Statistics, version 12.0 (SPSS Inc., Chicago, IL, USA).
Results
Baseline characteristics
Dosimetric data were obtained for 31,066 of the 33,931 CAs (92%), and for 25,356 of the 27,823 PCIs (91%) performed during 2010 in the 44 participating centres. Arterial access route was missing for 336 CAs and 382 PCIs, respectively. Therefore, among patients with dosimetric data and known arterial access route, radial access was used in 21,675/30,730 CAs (71%) and in 17,109/24,974 PCIs (69%), with significant differences across centres (the rate of radial access ranged from 1.7% to 94.1% for CA, and from 1.8% to 94.7% for PCI; both P < 0.0001). Right radial access was used in 33,005/38,784 transradial procedures (85%). Multiple arterial accesses were necessary in 2868/55,704 patients (5%) ( Supplementary Table ).
Baseline characteristics according to type of arterial access are shown in Table 1 . Compared with femoral access, transradial access was performed in younger patients (66 vs 68 years), and more frequently in male (73% vs 67%) and obese (26% vs 23%) patients. Emergency procedures and left ventriculography were more frequent in transfemoral procedures. Among PCIs, elective PCI using the femoral access was more frequent (9% vs 5%; P < 0.001).
| Overall ( n = 55,704) | Femoral access ( n = 16,920) | Radial access ( n = 38,784) | P | |
|---|---|---|---|---|
| Patients | ||||
| Age (years) | 67 [57–76] | 68 [58–77] | 66 [57–76] | < 0.001 |
| Male | 39,310/55,416 (71) | 11,280/16,856 (67) | 28,030/38,560 (73) | < 0.001 |
| BMI (kg/m 2 ) | 26.8 [24.3–30.0] | 26.6 [24.0–29.7] | 26.9 [24.2–30.1] | < 0.001 |
| BMI ≥ 30 kg/m 2 | 12,276/48,889 (25) | 3574/15,350 (23) | 8702/33,539 (26) | < 0.001 |
| Procedures | ||||
| CA | 30,730/55,704 (55) | 9055/16,920 (54) | 21,675/38,784 (56) | |
| Ad-hoc PCI a | 21,513/55,704 (39) | 6372/16,920 (38) | 15,141/38,784 (39) | |
| Elective PCI b | 3461/55,704 (6) | 1493/16,920 (9) | 1968/38,784 (5) | < 0.001 |
| Emergency | 7175/52,208 (14) | 2599/15,852 (16) | 4576/36,356 (13) | < 0.001 |
| Left ventriculography | 12,372/25,445 (48) | 4719/8926 (53) | 7653/16,519 (46) | < 0.001 |
| FFR | 961/47,248 (2) | 197/14,158 (1) | 764/33,090 (2) | < 0.001 |
| IVUS | 353/47,215 (1) | 89/14,152 (1) | 264/33,063 (1) | 0.11 |
a PCI immediately following coronary angiography in the same procedure.
Radiation dose variables and arterial access
Variables related to the patient’s radiation dose according to arterial access type are shown in Table 2 . By univariate analysis, median fluoroscopy time was significantly higher in CAs and PCIs performed using radial versus femoral access (CA, 3.8 vs 3.5 minutes [ P < 0.001]; PCI, 10.4 vs 10.1 minutes [ P = 0.001]). In contrast, median KAP was significantly lower for radial versus femoral access (CA, 26.8 vs 28.1 Gy·cm 2 ; PCI: 55.6 vs 59.4 Gy·cm 2 ; both P < 0.001).
| CA | PCI | |||||
|---|---|---|---|---|---|---|
| Femoral access ( n = 9055) | Radial access ( n = 21,675) | P a | Femoral access ( n = 7865) | Radial access ( n = 17,109) | P a | |
| Univariate analysis | ||||||
| KAP (Gy·cm 2 ) | 28.1 [16.4–46.9] | 26.8 [15.1–44.5] | < 0.001 | 59.4 [34.6–99.9] | 55.6 [32.2–92.1] | < 0.001 |
| Fluoroscopy time (minutes) | 3.5 [2.1–6.5] | 3.8 [2.3–6.3] | < 0.001 | 10.1 [6.2–16.7] | 10.4 [6.9–16.0] | 0.001 |
| Number of frames | 601 [432–846] | 526 [360.5–727] | < 0.001 | 867 [599–1,245] | 808 [560–1,143] | 0.03 |
| Number of runs | 10 [8–13] | 9 [8–12] | < 0.001 | 20 [15–27] | 19 [14–26] | 0.03 |
| Contrast media volume (mL) | 105 [75–140] | 90 [68–120] | < 0.001 | 180 [130–246] | 160 [115–220] | < 0.001 |
| Multivariable analysis | ||||||
| Mean KAP b (Gy·cm 2 ) | 36.4 (35.5–37.2) | 35.6 (34.9–36.2) | 0.001 | 77.7 (75.5–79.9) | 72.5 (70.9–74.2) | 0.001 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
