Background

Antithrombotic therapy and percutaneous or surgical myocardial revascularization procedures represent the basis of hospital treatment for patients admitted with non-ST-segment elevation acute coronary syndrome (NSTEACS) [1, 2]. However, the desired reduction of ischemic event recurrence is accompanied by an increased incidence of bleeding [3, 4]. Initially tolerated as an inherent complication of anti-ischemic treatment, major bleeding is now recognized as a predictor of mortality and adverse ischemic outcomes [5]. In fact, NSTEACS patients who develop major bleeding have a two- to ten-fold increase in mortality after a one-year follow-up [6, 7]. In the Acute Catheterization and Urgent Intervention Triage Strategy (ACUITY) trial [8] involving 13,819 NSTEACS patients submitted to an early invasive strategy and randomized for antithrombotic treatment with unfractionated heparin (UFH) plus IIb-IIIa glycoprotein receptor inhibitor (GPI), bivalirudin plus GPI or bivalirudin alone, the prognostic impact of major bleeding was comparable to that of acute myocardial infarction (AMI) in subsequent mortality (11.7% versus 9.1%, respectively) [9]. Randomized clinical trials involving NSTEACS patients show major bleeding rates varying from 2% to 5% [5, 8], approaching the prevalence of refractory ischemia, AMI or deaths observed in these studies. A publication of the National Cardiovascular Data Registry Acute Coronary Treatment and Intervention Outcomes Network Registry Get with the Guidelines (NCDR ACTION Registry-GWTG) [10], a representative registry of real world practice, has evaluated 72,699 unselected patients with NSTEACS and 48,943 patients with ST-segment-elevation AMI evaluated at 360 North American hospitals between January 2007 and June 2009. Authors have reported a major bleeding rate of approximately 9% among NSTEACS patients and 12% among those with ST-segment-elevation AMI, primarily influenced by the presence of comorbidities such as older age, female gender, chronic renal failure, as well as the use of invasive techniques [9, 11].

Bleeding and vascular complications related to arterial access

Because arterial puncture followed by sheath insertion using the modified Seldinger technique [12] has become the standard method used to perform invasive cardiovascular procedures, vascular access has become a major site of bleeding complications [13]. According to the Global Registry of Acute Coronary Events (GRACE), the most frequent bleeding sites were gastrointestinal (31.5%) and those related to vascular access (23.8%), with the latter being more prevalent among patients submitted to invasive strategies [11]. In a joint analysis of 17,393 acute coronary syndrome (ACS) patients submitted to percutaneous coronary intervention (PCI) and included in the studies Randomized Evaluation in PCI Linking Angiomax to Reduced Clinical Events (REPLACE) - 2, ACUITY and Harmonizing Outcomes with RevascularIZatiON and Stents in Acute Myocardial Infarction (HORIZONS-AMI), the bleeding prevalence by Thrombosis in Myocardial Infarction (TIMI) criteria was 5.3%, of which 2.1% (38.6%) were related to vascular access [14].

Strategies to prevent bleeding and vascular complications

Radial technique

Among the strategies to decrease vascular complications after invasive coronary procedures, the radial approach is an established one [15, 16]. Although it has been adopted by only a few centers, it provides more comfort to patients, allows early ambulation, decreases hospital stay and shows less vascular puncture site complications [17]. In a comparative randomized trial between radial and femoral techniques involving 7,021 ACS patients submitted to invasive techniques, both techniques were shown to be safe and effective for PCI, with similar incidences of death, AMI, stroke and major bleeding at 30 days (3.7% versus 4.0%; P = 0.50) [18]. However, the radial approach displayed significantly decreased vascular complications, including pseudoaneurysm, large hematomas, arterio-venous fistula and limb ischemia requiring surgical intervention (1.4% versus 3.7%; P < 0.0001).

Given the small number of training centers that use the radial technique, the uncertainties concerning the learning curve which would be associated with a higher failure rate and more radiological exposure, and the lack of large-scale multicenter studies that can reproduce the excellent results obtained by highly experienced centers, the femoral approach remains the most popular for invasive coronary procedures [19].

Percutaneous vascular closure devices

Although the real prognostic significance of preventing vascular access complications and minor bleedings has not yet been established, the occurrence of vascular access complications is associated with poor patient adherence to the antiplatelet therapy after hospital discharge and is a major and known predictor of ischemic complication recurrence [20]. Strategies to decrease femoral access vascular complications have been frequently evaluated and implemented, such as the use of smaller diameter endovascular devices, early arterial sheath removal, fluoroscopy or ultrasound-guided femoral puncture and the choice of antithrombotic agents with better safety profiles [8, 21–23].

Since 1995, percutaneous femoral vascular closure devices (VCD) were introduced to decrease vascular complications, hemostasis and ambulation times of patients submitted to invasive procedures by femoral access. Although these devices have been rapidly incorporated into interventionist practice, they have shown conflicting results regarding their safety and efficacy, motivating a class III recommendation as a strategy to decrease vascular complications in a recent position taken by the American Heart Association [24].

Three meta-analyses that compared VCD and manual compression from 2004 have reported conflicting results regarding their safety. Koreny et al. [25] grouped data from 30 randomized trials involving 4,000 patients and showed a decrease of 17 minutes to obtain hemostasis with VCD, at the expense of a non-significant increase in hematomas, bleeding, arterio-venous fistulas and pseudoaneurysms. Nikolsky et al. [26] identified 30 randomized and observational studies in 37,066 patients where VCD were associated with increased vascular complications compared with manual compression, particularly using the VasoSeal (Datascope Corp., Montvale, NJ) device. Conversely, Vaitkus et al. [27] described in a meta-analysis involving 15 randomized trials and 5,084 patients a decrease in vascular complications favoring the new technology, particularly when using AngioSeal (St. Jude Medical, St. Paul, MN) and Perclose (Abbott Vascular, Redwood City, CA) devices, an observation that has been validated by results of large multicenter observational registries published subsequently [28, 29].

In summary, different VCD, although not demonstrating a class effect, clearly provide more comfort to patients, decreasing hemostasis and bed rest time. However, the inconsistency of data proving their safety limits their routine adoption as a strategy to prevent vascular complications, requiring evidence through adequately designed randomized studies. The hypothesis of this investigation is that among NSTEACS patients submitted to an early invasive strategy and randomized to the femoral or radial approach, the percutaneous femoral vascular closure device would decrease the prevalence of vascular complications at the puncture site, thus fulfilling the non-inferiority criteria compared with radial access.

Methods

The ARISE study

Design

ARISE is a national, multicenter, non-inferiority randomized clinical trial comparing the radial versus femoral approach using VCD to decrease vascular complications related to an arterial puncture site among NSTEACS patients submitted to an early invasive strategy (Figure 1). The primary outcome will be evaluated from randomization until 30 days after the invasive coronary procedure.

Figure 1
figure 1

ARISE study design: a randomized trial of the radial versus femoral approach using AngioSeal in non-ST-segment elevation acute coronary syndrome (NSTEACS) patients.

Full size image

Outcomes

Primary outcome

Vascular complications at the arterial puncture site 30 days after the procedure (vascular complications at the arterial puncture site include major bleeding, retroperitoneal hematoma, compartment syndrome, hematoma ≥ 5 cm, pseudoaneurysm, arterio-venous fistula, infection, limb ischemia, arterial occlusion including asymptomatic occlusion, adjacent nerve injury or the need for vascular surgical repair).

Secondary outcomes

Individual components of the primary outcome, hematoma < 5 cm, major bleeding unrelated to the puncture site or myocardial revascularization surgery, device success and crossover rate; cardiovascular death, AMI or stroke at 12 months of evolution.

Study outcome definitions are listed in Table 1. A committee of clinicians will adjudicate all primary outcomes.

Table 1 Outcome definitions
Full size table

Patient population

Patients with NSTEACS will be managed with an invasive approach if the following pre-requisites are met: (i) palpable radial artery with normal Allen or oximetry tests, (ii) familiarity of the operator with the radial (≥ 100 transradial coronary procedures per year with a femoral crossover rate ≤ 4%) and femoral techniques using VCD (≥ 50 transfemoral coronary procedures per year with AngioSeal) and (iii) agreement of the operator to use the access route determined by the randomization process (Table 2).

Table 2 Eligibility criteria
Full size table

Procedures

Patients admitted with NSTEACS who are scheduled for early invasive stratification by coronary angiography followed by PCI, when applicable, will be evaluated in terms of feasibility of both radial and femoral access procedures. After the evaluation, with the patient meeting all inclusion criteria and in the absence of exclusion criteria, the patient may be included in the study after signing the free and informed consent term.

Patients will be randomized for the radial or femoral technique with VCD using a randomized sequence obtained by computer algorithms and maintained in individual, opaque and closed envelopes to conceal the allocation process.

Both radial and femoral coronary angiography will be performed using the Judkins technique and 6-French diameter sheaths and pre-molded catheters for selective catheterization of the left and right coronary arteries, with the possibility to change the diameter of the devices at the operator’s discretion. PCI will be indicated when a lesion presumably responsible for the adjacent clinical event is identified, with a stenosis diameter severity ≥ 70% and a high probability of angiographic success; PCI is ideally performed immediately after coronary angiography and left ventriculography. Patients with multiarterial coronary disease will be submitted to PCI after agreement among the clinical cardiologist, interventionist and heart surgeon. Procedures will be performed according to recommendations and provisions of current guidelines.

Arterial homeostasis

Radial approach

The TR Band device (Terumo Medical Corporation, Tokyo, JP) will be applied to obtain homeostasis according to a previously validated protocol [31, 32]. Immediately after procedure completion, the sheath is initially pulled by approximately 2 cm. The device is applied to the patient with the green marker (located in the center of the larger balloon) positioned exactly at the puncture hole to aid in the location, visualization and control of possible bleeding. The balloon is inflated with an adequate syringe injecting 15 mL of air with simultaneous and total sheath removal, resulting in the absence of active bleeding. From the fourth hour and at each subsequent hour (fifth and sixth hours), 5 mL of air is slowly removed, maintaining the balloon connected to the syringe and controlling the plunger with the thumb. If bleeding occurs during any stage of device removal, the volume of air needed for homeostasis is again injected, repeating the process 60 minutes later. If device failure occurs, homeostasis will be obtained using a compressive dressing with a porous elastic adhesive bandage (Tensoplast, Smith & Nephew, London, UK).

Femoral approach

The AngioSeal VCD, comprising an absorbable collagen sponge and absorbable polymer anchor with polylactic and polyglycolic acid connected by an absorbable self-adjustable suture, will be used for hemostasis. The device seals the arteriotomy hole between its two major components, the anchor and collagen sponge. Hemostasis is primarily obtained through mechanical means that is supplemented by collagen platelet aggregation-inducing properties. The device will be released according to the manufacturer’s recommendations. First, an insertion introducer together with a femoral artery arteriotomy hole locator will be positioned using a 0.035-inch-guidewire. Once blood reflow by the distal edge of the set is confirmed, the guidewire and insertion introducer are removed, keeping the arteriotomy locator in position. Next, the repair and hemostasis device are inserted through the locator, exposing the anchor in the intraluminal space. The retreat of this second set places the anchor against the internal puncture hole. Maintaining the retreat, the collagen sponge is then released in the external puncture hole at the same time that the anchor is sustained, resulting in effective and safe hemostasis by preventing intra-arterial collagen release. Total set removal is achieved using an absorbable suture wrapped in a plastic tube to manually compact the collagen sponge against the arterial wall. After the appearance of an opaque mark, the suture is cut, the compacting tube exerting compression may be removed, and the remaining suture is cut close to the skin. If the device fails, homeostasis will be obtained by manual compression. Patients will be allowed to walk immediately after the radial procedure, and one hour after bed rest in the supine position after the femoral procedure with AngioSeal.

Table 3 illustrates possible antithrombotic treatment. After successful PCI, anticoagulant therapy will be withdrawn. Drugs such as angiotensin-converting enzyme inhibitors or angiotensin receptor blockers, beta blockers and statins will be prescribed according to current secondary prevention guidelines.

Table 3 Adjunct antithrombotic therapy
Full size table

Electrocardiogram (ECG), CK-MB and/or troponin, glucose, creatinine, sodium, potassium, blood count and coagulation tests shall be obtained before patient randomization. Biomarkers (CK-MB and/or troponin) shall be checked between 12 and 24 hours after the procedure. ECG shall be performed soon after the procedure and within 12 to 24 hours after the procedure or when a new ischemic event is suspected. Hemoglobin and hematocrit dosage will be required in the presence of vascular complications or bleeding. Vascular and systemic complications related to the arterial vascular access will be evaluated in the interval between procedure completion, hospital stay and the first 30 days of evolution, through a scheduled visit. Other outcomes will be considered until the 30 days of evolution, with late follow-up by a visit or telephone call regarding cardiovascular death, AMI or stroke, at 12 months.

Statistical analysis

The primary analysis of the study is a non-inferiority comparison between the radial approach and femoral approach using VCD to decrease vascular complications at the arterial puncture site, in all randomized patients, based on the intention-to-treat principle.

Estimating a vascular complication rate pre-specified in the primary outcome of approximately 3% for the radial technique [19, 32] and 12% for the femoral technique [26–31] and determining a non-inferiority margin of 0.15 based on historical data [18, 19] with an alpha level of 0.05 and a beta level of 0.10, the minimum estimated sample size per group was established as 97 individuals. Twenty additional patients may be added to the final population to correct any subsequent loss of follow-up. This non-inferiority margin was derived from a trial that demonstrate the benefit of radial approach on the reduction of major vascular complications (hazard ratio, 0.37; 95% confidence interval, 0.27 to 0.52) as compared with femoral approach [18]. The non-inferiority margin of 0.15 was chosen in order to avoid a loss of greater than half the lower bound of the 95% confidence interval (0.27). The non-inferiority of the femoral approach with VCD will be declared if the lower confidence interval limit of 95% of the difference of both techniques does not include the specified inferiority margin value.

Categorical data will be presented as frequencies and group percentage and will be compared by chi-squared or Fisher’s exact tests. Continuous variables will be expressed as means and standard deviation and will be compared by Student’s t- test. Statistical analyses will be performed using Statistical Package for the Social Sciences for Windows (version 16.0; SPSS, Chicago, IL, USA).

The study was approved by the Dante Pazzanese Institute of Cardiology ethical committee, predicting and validating the Santa Casa de Marília as a co-participating institution. No extramural funding will be used to support this work. The authors are solely responsible for the design and conduct of this study, all study analyses and drafting and editing of the paper.

Results and discussion

Study status

Enrollment was initiated in September 2012, and until October 2013 91 patients were included. The inclusion phase is expected to last until the second half of 2014.

Conclusions

Despite the proven efficacy of the radial approach in reducing vascular complications at the puncture site, the femoral approach remains the preferred technique at many centers worldwide. The ARISE trial will help define the role of vascular closure devices as a bleeding avoidance strategy in patients with NSTEACS.