Abstract

Trial Design

The EXPERT trial was a randomized noninferiority trial comparing TRA and TFA for diagnostic CA (Fig. 1).

View Figure

• Download as Powerpoint
• Download Figure

Participants

Patients with suspected coronary artery disease and a clinical indication for CA were invited to participate. The inclusion criteria were indications for elective diagnostic CA, eligibility for either TRA or TFA, and age 21 years or older. Patients were excluded if they were scheduled for therapeutic procedures; underwent urgent or emergency procedures or simultaneous right-sided catheterization; had end-stage renal disease; or required additional, noncoronary angiography during CA.

Interventionalist Expertise

All procedures were performed by experienced interventional cardiologists who fulfilled the following criteria: (1) full training in both radial and femoral access for coronary diagnostic and therapeutic procedures, (2) completion of a minimum of 500 diagnostic procedures in the past 2 years, and (3) level III competency according to Society for Cardiovascular Angiography & Interventions (SCAI) guidelines.¹⁴ Level III competency indicates an ability to perform complex interventional procedures, even in patients with challenging limb anatomy (eg, chronic total occlusion, multivessel disease) or acute myocardial infarction.

Imaging Protocol

After screening was completed and patients provided written informed consent, they were randomly assigned to either TRA or TFA. In both groups, at least 9 pre-specified basic views per patient were required (5 runs in the left coronary artery, 3 in the right coronary artery, and 1 in the left ventricle). Images were obtained at various angulations, in the following sequence: left ventriculography (right anterior oblique [RAO] 30° with no caudal or cranial angulation), left coronary (RAO 20°/caudal 20°, anteroposterior [AP] caudal 20°, left anterior oblique [LAO] 40°/caudal 30°, LAO 40°/cranial 25°, AP cranial 40°), and right coronary (LAO 30°, LAO 30°/cranial 30°, RAO 30°). Additional views were allowed if they were necessary for an accurate diagnosis. Appropriate use of edge filters, collimators, and flat detectors close to the patient's chest was required. All procedures were performed using low-dose fluoroscopy guidance (maximum radiation exposure, 50 milligray [mGy]/min) at a capture rate of 15 frames/s. Coronary images were acquired at a magnification of 20 cm, and left ventricular images were acquired at a magnification of 25 cm. Operators wore complete radiation protection gear (ie, a lead apron, leaded glasses, and upper- and lower-body shields).

Although patients with certain conditions were excluded from the study, those with a history of coronary artery bypass graft (CABG) and severe valvular heart disease were eligible to participate. For patients with previous CABG who were assigned to TRA, the right or left radial artery could be used according to the operator's discretion. Imaging acquisition included the 9 basic angiographic views described above plus 2 additional recommended runs per graft. Procedures in patients with severe valvular heart disease were performed with the 9 prespecified views. For patients with severe aortic stenosis, for example, operators were encouraged to cross the valve, perform left ventriculography, and measure the transaortic pressure gradient.

Procedures

TRA. Patients were supine with their arms at their sides; a sterile drape was placed. Local anesthetic (3–5 mL of 2% lidocaine) was injected subcutaneously, 2 cm proximal to the styloid process, over the palpable radial artery. The artery was punctured with a 22-gauge needle, and a 0.035-inch guidewire was inserted into the vessel. A small skin incision was made, and a 5F or 6F radial sheath (Radifocus Introducer II Standard Kit A; Terumo Interventional Systems) was introduced. Nitroglycerin (100–300 μg) was administered to induce vasodilation, and a bolus of heparin (5,000 units) was given intravenously to all patients after the guidewire reached the ascending aorta. Either a 5F or 6F catheter was used for CA; the catheter shape was selected according to operator discretion. The sheath was removed immediately after the procedure, regardless of the patient's level of anticoagulation, and a compression dressing was applied. Four hours later, the compression bandage was removed. Patients were discharged from the hospital on the same day.

TFA. Femoral access was obtained using traditional technique. A sterile drape was placed, and local anesthetic (10 mL of 2% lidocaine) was injected subcutaneously into the inguinal area. Femoral puncture was performed primarily using palpation of the inguinal ligament for guidance; if needed, fluoroscopic guidance was used. All operators were advised to perform the anterior wall puncture technique only. Either 5F or 6F introducers (Avanti+; Cordis) and catheters were used according to the operator's discretion. The sheath was removed immediately after the end of the procedure, and manual compression was applied for hemostasis. No dedicated closure devices were used.

Sedation. For all patients (TRA and TFA groups), mild sedation with fentanyl and midazolam was allowed. The dose administered was at the operator's decision.

Definitions

The imaging protocol was considered complete when selective catheterization of the coronary arteries and the left ventricle was achieved. Puncture time was defined as the time from local anesthesia to introducer insertion, and the total procedure time as the time from local anesthesia until the last catheter was removed.

Outcomes

The primary end point was the patient radiation dose during CA. Patient radiation exposure was measured as cumulative air kerma (kinetic energy released per unit of mass) in mGy and dose area product (DAP) in mGy·cm². The secondary end points were operating physician radiation exposure and major cardiovascular events at 30 days. Physician exposure (in μSv) was assessed using an individual digital dosimeter (PM1621; Polimaster, Inc). The operating physician effective dose (ED) was determined according to the following formula: ED = (PD − BR) × CC, where PD is the procedure dose (μSv), BR is the background radiation (procedure time [s] × 0.00004 μSv/s), and CC is the conversion coefficient (1.01).

Follow-Up

Patients were followed up at 24 hours for vascular complications, bleeding (using Global Utilization of Streptokinase and Tissue plasminogen activator in Occluded arteries [GUSTO] criteria),¹⁵ and stroke. They were then followed up at 30 days for major cardiovascular events.

Sample Size and Statistical Considerations

The EXPERT trial was designed as a noninferiority study, to show that TRA does not cause significantly more radiation exposure than TFA. A prior nonrandomized study by the same researchers¹⁶ showed that diagnostic CA using TRA caused 39% more radiation exposure than CA using TFA (621.6 mGy vs 445.7 mGy). The EXPERT trial was based on the assumption that experienced interventional cardiologists could reduce this excess exposure; a sample size of 300 patients was calculated to provide 80% power to detect any difference greater than 20% (relative margin) between groups.

Continuous variables were reported as the mean (SD) for normally distributed variables and as the mean (25th–75th percentile range) for those not normally distributed; these data were compared using the Student t test or the Mann-Whitney U test, as appropriate. Categorical variables were reported as the absolute number and percentage and were compared using the Pearson χ² test or the Fisher exact test. Predictors of higher radiation exposure were identified using multiple regression analysis. SPSS for Windows, version 16.0 software (IBM) was used for all analyses. P ≤ .05 was considered statistically significant.

All analyses followed the intention-to-treat principle, that is, if a patient's CA procedure was converted from TRA to TFA (or vice versa), the patient was included in their original group for analysis.

Masking and Randomization

The operating physicians and patients could not be blinded to the intervention. However, the investigator who screened patients for the study and obtained informed consent was blinded to the treatment assignment. Numbered, sealed envelopes were used to randomly assign patients in a 1:1 ratio to the TRA and TFA groups. Operating physicians were informed about the assigned access method only when the patient was in the operating room, draped, and ready for the procedure.

Procedures

As shown in Table II, the right radial and right femoral arteries were the most used access sites for CA. Coronary artery disease severity, left ventricular function, and the frequency of technical difficulties were comparable between groups. The imaging protocol (coronary artery and left ventricle catheterization) was completed in 97.3% (146/150) of patients in the TRA group and 99.3% (148/149) of patients in the TFA group (P = .99). There were no differences in mean (SD) puncture time (TRA, 2.13 [1.8] vs TFA, 2.2 [4.9] min; P = .87), mean (SD) total procedure time (15.05 [4.8] vs 14.1 [4.3] min; P = .07), or mean (SD) fluoroscopy time (3.52 [2.18] vs 3.13 [2.43] min; P = .14). The number of projections used was also equivalent (TRA, 9.65 vs TFA, 9.68; P = .93). TRA required significantly less mean (SD) contrast volume (91.4 [21.2] mL vs 96.23 [23.7] mL; P = .05) and a smaller mean (SD) number of catheters (2.4 [0.3] vs 3.2 [0.6]; P < .001).

TABLE II. Patient and Procedure Characteristics

View Fullscreen

Primary and Secondary End Points

Table III summarizes the radiation exposure for patients (kerma and DAP) and physicians.

TABLE III. Patient and Physician Radiation Exposure

View Fullscreen

Patients With Previous CABG

In the TRA and TFA groups, 5.3% (8/150) and 5.4% (8/149) of patients had a history of CABG (P = .99). Procedural characteristics were similar between groups regarding the mean (SD) number of saphenous grafts per patient (2 [1] vs 2.5 [0.9]; P = .33), mean (SD) number of left internal mammary artery grafts per patient (1 vs 1, P = .99), mean (SD) number of angiographic views per procedure (12.6 [1.5] vs 13.1 [1.4]; P = .51), and mean (SD) fluoroscopy time (6.77 [3.75] vs 5.68 [2.43] min; P = .51). The TRA and TFA groups had comparable mean (SD) patient kerma values (760.1 [429] vs 872.7 [201] mGy; P = .51) and mean (SD) physician doses (1 [0.5] vs 0.8 [0.3] μSv; P = .41). Likewise, the mean (SD) DAP did not differ significantly between groups (46,971.7 [29,259] vs 56,948.5 [17,552] mGy·cm²; P = .42).

Patients With Severe Valvular Heart Disease

Severe valvular heart disease was present in 11.3% (17/150) and 12.1% (18/149) of the TRA and TRA patients (P = .99). Mean (SD) patient radiation exposure (557.2 [198] vs 613.2 [153] mGy; P = .35), mean (SD) operator exposure (0.56 [0.3] vs 0.64 [0.2] μSv; P = .44), mean (SD) runs per procedure (9.94 [1.3] vs 9.67 [1.2]; P = .53), and mean (SD) fluoroscopy time (4.23 [2.51] vs 4.71 [2.9] min; P = .61) were similar between groups. However, mean (SD) DAP was significantly higher in the TFA group (32,737 [9,890] vs 41,428 [11,065] mGy·cm²; P = .02).

Follow-Up

There were no instances of in-hospital vascular complications, stroke, or bleeding (according to GUSTO criteria) in either group. At 30-day follow-up, 1 patient had experienced a stroke (not procedure related) in each group, and 2 patients in the TFA group (1.3%) experienced pseudoaneurysm. Almost all patients were completely asymptomatic: 99.3% in the TRA group and 98% in the TFA group (P = .24).

Predictors of Greater Radiation Exposure

On examination of clinical characteristics that have traditionally been associated with TRA failure or greater radiation exposure, there were no significant predictors of higher radiation dose (Fig. 2).

View Figure

• Download as Powerpoint
• Download Figure