A Randomized Comparison of Radial Artery Intimal Hyperplasia Following Distal vs. Proximal Transradial Access for Coronary Angiography: PRESERVE RADIAL Trial

Subjects, Study Design

The study was approved by the Inova Health System’s institutional review board, registered at clinicaltrials.gov (Registration number: NCT04801901) and supported by an investigator-initiated research grant from Boston Scientific Corporation (Maple Grove, MN). The study authors were solely responsible for the design and conduct of this study and drafting and editing of the manuscript. Statistical analysis was provided by the George Mason University Department of Biostatistics. The study sample was drawn from consecutive patients undergoing elective and non-emergent CAG and/or PCI at Inova Fairfax Medical Campus. Eligibility criteria included: age ≥ 18 years, non-emergent CAG and/or PCI in the setting of symptomatic ischemic heart disease (stable angina, unstable angina, and non-ST segment elevation myocardial infarction). Study subjects had to agree to return for 90-day post-procedural RA ultrasound and completion of BORG and DASH questionnaires for hand pain and functional assessments.^6,7 Exclusion criteria included: confirmed pregnancy, uncorrected bleeding disorders, inability to take antiplatelet therapy, prior cannulation of both RA’s, acute ST-segment myocardial infarction, cardiogenic shock, and known hypersensitivity to stainless steel, platinum, chromium, nickel, molybdenum, or everolimus. In patients with prior RA cannulation, the contralateral vessel was accessed. Patients were randomized 1:1 to dTRA or fTRA using sequentially numbered, opaque sealed envelopes. Permuted block randomization was used to stratify for patient sex to ensure equal distribution of women in each group.⁸ Patient study flow is shown in Figure 1.

• Download figure
• Open in new tab

Figure 1.

PRESERVE Radial Consort Flow Diagram and Study Design. Abbreviations: dTRA = distal transradial access; FRA = forearm radial artery; fTRA= Forearm transradial access; IMT = intimal medial thickness; US = ultrasound.

Ultrahigh Resolution Ultrasound Imaging Technique

Use of the right or left RA for was left to physician discretion. Distal and FRA segments were evaluated at baseline and 90 days using FUJIFILM VisualSonics 55-MHz ultrahigh resolution duplex ultrasound. With spatial and lateral resolutions of <40 μm and < 80 μm, respectively, this device distinguishes between the anatomic layers of the RA.⁹ In patients without contraindications, nitroglycerine (0.4 mg) was administered 3-5 minutes prior to ultrasound to provide optimal arterial vasodilation.¹⁰ Patients underwent ultrasound imaging of both the distal and FRA segments at baseline and 90 days. RA IMT was measured using a technique that has been previously reported.¹⁰ For assessment of the FRA, the ultrasound probe was placed over the FRA with imaging beginning at the distal end of the radius bone and progressing proximally. To measure the average IMT over an appropriate length of the vessel, a 3 cm length of the FRA was assessed, starting from centering the probe over the distal radius and moving proximally. Three successive IMT measurements were recorded within each of 3 one centimeter long segments. The numerical mean of all measurements was recorded as the final IMT. For assessment of the dTRA segment, the ultrasound probe was positioned over the triangular depression on the dorsum of the hand bordered by the extensor pollicis brevis and the abductor pollicis longus tendons laterally and by the extensor pollicis longus tendon medially. In a similar manner, three 1 cm segments of vessel were also acquired and the IMT recorded as the numerical mean of all measurements. The IMT was measured as the distance between the intima and media-adventitia interfaces (Figure 2). Final measurements of IMT were validated by an interventional cardiologist (WB) blinded to treatment allocation. Other measurements recorded included RA cross-sectional vessel and lumen diameters. Vessel diameter (i.e., inter-adventitial diameter) was measured as the distance between the media-adventitia interfaces of the anterior and posterior walls of the RA. Lumen diameter was measured as the cross-sectional distance between the anterior and posterior intima. We also assessed for signs of vessel trauma, including limited access site intimal tears, medial dissections, and RA thrombosis, occlusion and/or pseudoaneurysms.¹⁰

• Download figure
• Open in new tab

Figure 2.

Two-dimensional ultrahigh resolution ultrasound of radial arterial intimal medial thickness (IMT) using a 55 MHz probe. IMT was measured as the distance from the intima to the outer edge of the media.

Coronary Angiography and PCI Technique

In patients undergoing dTRA, the right or left hand was positioned with the thumb flexed underneath the other four digits, and the hand positioned above the right groin. After sterile preparation and administration of 2% lidocaine subcutaneously, arterial puncture was performed under ultrasound guidance using a single anterior wall puncture technique. Once access was confirmed, a 0.014”x190 cm hydrophilic Hi-Torque Whisper MS guidewire (Abbott®, Chicago, IL) was advanced under fluoroscopic guidance to the level of the elbow, at which time the access needle was removed, and the coronary wire served as a rail for sheath insertion. insertion of a 6/7 Fr Glidesheath Slender. In patients undergoing fTRA, the hand was supinated and positioned on an arm board directly adjacent to the patient’s ipsilateral hip. Arterial access was obtained using ultrasound guidance and either a single or double-wall puncture (according to physician preference). A 6/7 Fr Glidesheath Slender (Terumo®, Somerset, NJ) was used in both cohorts given its hydrophilic coating and large inner diameter, allowing for advancement of 6 or 7 Fr guide catheters for PCI. All patients received a RA antispasmodic cocktail consisting of 5.0 mg verapamil, 200 mcg nitroglycerine and 5000 units of unfractionated heparin administered through the sheath.¹¹ If PCI was performed, additional I.V. heparin was administered to achieve an activated clotting time (ACT) of 250-300 seconds, or 200-250 seconds if a glyocoprotein IIb-IIIa inhibitor was given. Patients undergoing PCI were considered for a SYNERGY or Promus Elite (Boston Scientific, Maple Grove, MN) drug eluting stent; however final stent selection was according to physician discretion. All patients undergoing PCI received dual antiplatelet therapy with oral aspirin (81 mg) and a P2Y12 platelet inhibitor bolus (600 mg clopidogrel, 60 mg prasugrel or 180 mg ticagrelor).

Radial Artery Hemostasis Protocol

Following CAG and/or PCI, RA hemostasis was achieved using a standardized patent hemostasis protocol.¹¹ In the fTRA cohort, a trained cardiac catheterization laboratory cardiovascular technician or nurse applied a TR Band radial compression device (Terumo®, Somerset, NJ) to the wrist over the access site with the green box on the band placed proximal to the arteriotomy. As the sheath was slowly withdrawn, 15 mL of air was injected into the one-way valve on the TR band, a pulse oximeter placed over the thumb and the plethysmography waveform observed. While performing manual occlusion of the ipsilateral ulnar artery, the hemoband was tightened until the plethysmography waveform was obliterated. The TR band was then loosened 1 mL at a time until the waveform first returned and was maintained, thus confirming patent hemostasis. In the dTRA cohort, a Safeguard Compression Device (Merit Medical®, Jordan, UT) 1-2 mm was applied proximal to the arteriotomy, and 3 mL of air was inflated into the pillow compartment while the sheath was withdrawn. In both patient cohorts, hemostasis was maintained for at least 2 hours following diagnostic CAG and at least 4 hours following PCI.

Statistical Analyses

The primary study endpoint was ipsilateral forearm IMT measured at 90 days. We also compared the change in FRA IMT at 90 days vs. baseline between the dTRA and fTRA cohorts. An intention to treat analysis was used and testing for superiority between the two groups. The null hypothesis assumed no between group difference in FRA IMT at 90 days. From previous studies, we anticipated that 90-day FRA IMT would be approximately 0.34 ± 0.08 mm for the control group (fTRA).¹⁰ Assuming similar standard deviation and a two-sided test with type 1 error of 0.05, we estimated that a sample size of 56 would provide 90% power to detect a 0.07 mm (20%) difference in 90 day IMT.¹⁰ To account for 10% patient drop-out or lost to follow-up, an additional 3 patients per group would be required. Summary statistics were presented as a mean ± standard deviation (SD), or frequency and percent, where deemed appropriate. Comparisons were made via Student’s t test for continuous variables and chi-square, or Fisher’s exact for discrete variables. Statistical significance was defined as a p-value < 0.05 using two-sided tests. Multivariable regression analysis was performed to examine for a relationship between baseline clinical demographics and 90-day FRA IMT. Variables included in the model were those historically associated with incidence of RAO: age, female gender, tobacco use, diabetes, and chronic renal insufficiency.^11,12 All analyses were performed using R (4.0.2) software for statistical computing.

Secondary Endpoints

Clinical and procedural data were recorded. Secondary endpoints included: number of cannulation attempts to obtain access, TRA failure necessitating cross-over to either an alternative RA or femoral arterial access, total procedure time, radiation exposure (cumulative air kerma and fluoroscopy time), contrast utilization, duration of hemostasis, and incidence of ≥ Class III hematomas, as defined by the EASY criteria (Early Discharge after Transradial Stenting of CoronarY Arteries Study).¹³ In patients undergoing PCI, procedural outcomes and complications were also recorded, including successful PCI (defined as <30% residual stenosis with TIMI 3 flow), coronary perforation, in-hospital death, myocardial infarction and urgent revascularization within 24 hours.g¹⁴ Other secondary endpoints included patient perceived pain and function of the ipsilateral hand at 90 days using the Borg and DASH (Disabilities of the Arm, Shoulders, and Hand) scales, respectively^6,7. The study consort flow diagram is shown in Figure 1.

Patient and Procedural Characteristics

From October 1, 2021, to March 31, 2022, 64 consecutive patients were randomized: 33 in the dTRA group and 31 in the fTRA cohort. Two patients in the dTRA and 4 in the fTRA cohort were lost-to-follow up. Baseline clinical and procedural characteristics are shown in Tables 1 and 2 and patient study flow in Figure 1. For the overall sample, mean age was 67.5 years, 70% were male, 83% were white, 36% diabetic and 7% had chronic renal failure. There were no differences in baseline clinical characteristics between the dTRA and fTRA cohorts (Table 1). Mean baseline left ventricular ejection fraction was 57%, 80% of patients presented with stable ischemic heart disease and 25% had undergone prior PCI. Seventy two percent of all patients (n=46) underwent right arm TRA. No between group differences were noted in the number of cannulation attempts required for RA access (1.2 vs 1.0, respectively; p=0.11), incidence of RA spasm (0% vs 6.5%; p=0.23), access site cross-over (3.0% vs 0%; p=1.0), radiation exposure (0.6 Gy vs 0.5 Gy; p=0.48), contrast utilization (79 cc vs 93 cc; p=0.34), and post-procedural hemostasis time (157 mins vs 161 mins; p=0.86). Thirty-four percent of all patients (n=22) underwent PCI (33% of dTRA patients vs. 36% of fTRA patients; p =1.0), and 19% had adjunctive atherectomy performed (26% of dTRA PCIs vs. 11% of rTRA PCIs; p=0.41). All PCIs were successful with no coronary dissections, perforations, or acute stent thromboses. There were no post-procedural RA access site hematomas (EASY Class ≥3) in either group.

Baseline Ultrahigh Resolution Ultrasound Results

Baseline RA ultrasound measurements are shown in Table 3. There were no differences in baseline distal RA vessel diameter (2.80 mm vs. 2.92mm, respectively; p=0.18), distal RA luminal diameter (2.03 mm vs 2.19 mm; p=0.08) and distal RA IMT (0.31 mm vs. 0.31 mm; p=0.99). Similarly, there were no differences in baseline forearm RA vessel diameter (3.31 mm vs. 3.35 mm; p=0.73), forearm RA luminal diameter, 2.56 vs. 2.64 mm; p=0.47) and forearm RA IMT (0.31 mm vs 0.31 mm; p=0.97).

90 Day Ultrahigh Resolution Ultrasound and Functional Assessments

Ninety-day ultrasound findings, and Borg and DASH scale assessments are shown in Table 4. Complete data were available for the 58 patients (31 dTRA cohort and 27 fTRA) who completed 90-day ultrasound measurements. There were no differences in 90-day FRA IMT (0.37 mm vs 0.38 mm; p =0.73), change in IMT (+0.07 mm vs +0.07 mm; p=0.61), vessel diameter (3.04 vs. 2.95 mm; p=0.48), change in vessel diameter (-0.29 mm vs. -0.39 mm; p=0.13), luminal diameter (2.32 mm vs. 2.34 mm; p=0.85), and change in luminal diameter (-0.25 mm vs -0.32 mm, p=0.27) (Figure 3). No access site intimal tears, dissections, RA occlusions or RA pseudoaneurysms were noted at 90 days and there were no differences in 90-day Borg pain scale (12 vs 11; p=0.24) or DASH scores (6 vs 8; p=0.46).

• Download figure
• Open in new tab

Figure 3.

Box and whisker plots depicting the change in forearm radial artery intimal medial thickness from baseline to 90 days.

Ultrasound of the distal RA showed significant differences between groups. At 90 days, patients in the dTRA group showed greater distal RA IMT (0.38 mm vs 0.31 mm; p < 0.001), larger increase in dTRA IMT (+0.07 vs. 0.00; p<0.001), smaller distal RA vessel diameter (2.47mm vs. 2.87mm; p<0.001), greater reduction in distal RA vessel diameter (-0.33 mm vs. -0.03 mm; p<0.001), smaller distal RA luminal diameter (1.77 mm vs 2.20 mm; p < 0.001) and greater reduction in distal RA luminal diameter (-0.26 mm vs. 0.00 mm; p<0.001) compared to those undergoing fTRA. Multivariable regression analysis identified age and diabetes mellitus as the only independent predictors of 90-day forearm IMT. With every 1-year increase in age, there was a 1.5x10^-3 mm increase in forearm RA IMT at 90 days (95% CI: 0.78 x 10^-3 to 0.28 x 10^-2; p<0.01). Similarly, diabetes mellitus was associated with a 2.3 x 10^-2 mm increase in forearm RA IMT at 90 days (95% CI: 0.13 x 10^-2 to 0.55 x 10^-1; p=0.04).