The use of glycoprotein IIb/IIIa receptor inhibitors (GPIs) in high-risk patients with acute coronary syndromes has been associated with reductions in ischemic events but increases in bleeding complications. The role of GPIs in patients who undergo primary percutaneous coronary intervention (PCI) by the transradial approach (TRA) is not well studied. The aim of this post hoc analysis from the randomized prospective Acute Myocardial Infarction Treated With Primary Angioplasty and Intravenous Enoxaparin or Unfractionated Heparin to Lower Ischemic and Bleeding Events at Short- and Long-Term Follow-Up (ATOLL) trial was to assess the safety and efficacy of GPIs in primary PCI performed using the TRA. A total of 910 patients were enrolled in ATOLL; 522 patients (67%) underwent PCI using the TRA. Two comparative analyses were performed. First, patients who underwent PCI using the TRA who received GPIs were compared with those who did not receive GPIs. Second, patients who underwent PCI using the TRA who received GPIs were compared with those who underwent PCI using a nonradial route and received GPIs. Composite end points of net clinical benefit, ischemic outcomes, and safety consisting of bleeding and transfusion at 1 month were analyzed. A propensity score was constructed, and weight adjustment were made for variables, including but not limited to age, weight, gender, renal function, concomitant use of other medications, Killip class, and medical history, when analyzing the end points. There was no significant difference in net clinical benefit or ischemic outcomes between either TRA patients with versus without GPIs or TRA patients with GPIs versus non-TRA patients with GPIs. Additionally, there were significantly fewer major bleeding events and blood transfusions in TRA patients with GPIs compared with non-TRA patients with GPIs. In conclusion, the addition of GPIs in the setting of primary PCI using the TRA was not associated with bleeding liability. The use of GPIs with TRA was associated with safer outcomes than using GPIs with a nontransradial approach. This study was limited in that it was a nonrandomized retrospective analysis.
Glycoprotein IIb/IIIa receptor inhibitors (GPIs) are widely used as potent adjuvant antiplatelet therapy to prevent platelet-rich clot formation in the spectrum of acute coronary syndromes (ACS) treated with percutaneous coronary intervention (PCI). Reductions in ischemic events in patients with non–ST-segment elevation myocardial infarction (STEMI) ACS or those with STEMI ACS come at the price of more bleeding complications. Despite the lack of a clear causal relation, major bleeding encountered during ACS is associated with an increase in adverse outcomes. To reduce the incidence of bleeding, various strategies have been adopted, including the performance of cardiac catheterization by a transradial approach (TRA). Our knowledge of how GPIs might affect the precarious risk-benefit balance of reducing ischemic versus bleeding complications in patients with STEMIs treated with the TRA is very limited. To better understand the role of GPIs in patients who undergo primary PCI using the TRA, we conducted a post hoc analysis of the Acute Myocardial Infarction Treated With Primary Angioplasty and Intravenous Enoxaparin or Unfractionated Heparin to Lower Ischemic and Bleeding Events at Short- and Long-Term Follow-Up (ATOLL) trial. ATOLL was a prospective, multicenter, randomized trial assessing the efficacy and safety of intravenous low–molecular weight heparin versus unfractionated heparin in patients with STEMI treated with primary PCI. More than 2/3 of the patients in ATOLL were treated with PCI by the TRA and adjunctive GPIs. We conducted 2 separate subgroup analyses. First, we compared clinical outcomes in patients who underwent the TRA with GPIs versus the TRA without GPIs. Second, we compared clinical outcomes in TRA patients with GPIs versus non-TRA (mostly transfemoral) patients with GPIs.
Methods
Details of the ATOLL trial were presented in the original publication. In short, ATOLL was an international, randomized, open-label trial evaluating intravenous enoxaparin (0.5 mg/kg) versus intravenous unfractionated heparin in patients who underwent primary PCI for STEMI. Patients with STEMIs were eligible to enter ATOLL if they were >17 years of age (without an upper age limit) and had indications for primary PCI <12 hours after symptom onset. Patients presenting from 12 to 24 hours of symptom onset with persistent ischemic symptoms or persistent or recurrent ST-segment elevation on electrocardiography, or both, and indications for primary PCI were also eligible, as were patients with shock or cardiac arrest (<10 minutes) in the setting of STEMI. The use of concomitant drugs, including GPIs, was at the discretion of the treating clinicians. Patients who received anticoagulants of any type before randomization were excluded. Other major exclusion criteria were the administration of thrombolytic agents for the present episode, a short life expectancy, childbearing potential, and known contraindications to treatment with antithrombotic agents.
All patients received aspirin (75 to 500 mg/day), thienopyridines, and GPIs according to local practice. No changes to the doses of enoxaparin were made with or without the use of GPIs, whereas dose adjustments were made to unfractionated heparin depending on the receipt of GPIs on the basis of current guidelines. Arterial closure devices were allowed after femoral access. All technical aspects concerning mechanical reperfusion, thrombectomy, choice of stents, or hemodynamic support were left to the discretion of the treating clinicians. Clinical follow-up took place at 30 days (within 2 days).
The primary end point of the ATOLL trial was the occurrence of the composite end point of death, complication of myocardial infarction, procedure failure, or major bleeding. Major bleeding unrelated to coronary artery bypass grafting during the hospital stay was defined according to the Safety and Efficacy of Enoxaparin in PCI Patients, an International Randomized Evaluation (STEEPLE) definition. The main secondary efficacy end point was the composite of any death, recurrent ACS, or urgent revascularization and was the first end point tested after the primary end point. The main safety objective was major bleeding unrelated to coronary artery bypass grafting during the hospital stay.
Another secondary safety objective was the composite of major and minor bleeding during the hospital stay (STEEPLE definitions). The net clinical benefit end point was prespecified as the combination of death, complication of myocardial infarction, or major bleeding. Detailed definition of each end point are available in the original ATOLL publication. All clinical events were adjudicated by an independent clinical events committee that was unaware of the treatment assignments.
For all analyses, continuous variables were compared using Student’s t tests or Mann-Whitney tests and are summarized as mean ± SD or as median (interquartile range) in case of a non-Gaussian distribution. All qualitative variables were compared using chi-square tests for frequency comparisons and Fisher’s exact probability tests when the criteria for chi-square analysis were not fulfilled and are expressed as frequencies and percentages. Relative risk and 95% confidence intervals were calculated for end points.
For the comparison of groups of patients treated with the TRA with GPIs versus the TRA without GPIs and those treated with the TRA with GPIs versus non-TRA patients with GPI, we built a propensity score on the basis of baseline characteristics of the patients to adjust the estimators, taking into account the nonequiprobability of the 2 groups and the possible strong relation between patient characteristics and the administration of vascular access or administration of GPIs. Adjustment of treatment effects estimators by propensity score used the inverse weight method.
Time to events was compared using either log-rank tests for per protocol population or score tests, taking into account propensity score. All analyses were performed using the SAS version 9.2 (SAS Institute Inc., Cary, North Carolina).
Results
A total of 910 patients were enrolled in the ATOLL study, of whom 614 (67%) underwent PCI using the TRA and 296 (33%) using a nontransradial approach. Seven hundred twenty-nine of the 910 patients (80%) in the ATOLL trial were treated with GPIs. Of 614 patients who underwent PCI by the TRA, 522 (85%) were treated with GPIs, and 207 patients (70%) in the non-TRA group received GPIs ( Figure 1 ).
The baseline clinical and procedural characteristics of the 614 TRA patients treated with or without GPIs are listed in Table 1 . A few baseline and procedural variables were significantly different. Table 2 lists the relative risk for the primary safety and efficacy end points after the construction of a propensity score and weight adjustment when analyzing these end points. No significant difference was observed in patients treated with the TRA with GPIs compared with TRA patients without GPIs in the primary composite end point of death, complication of myocardial infarction, procedure failure, or major bleeding. Nor was a significant difference observed in the main secondary efficacy end point, a composite of death, recurrent ACS, or urgent revascularization. Although none of the efficacy end points reached statistical significance in favor of the TRA with GPIs, numerically, there was a consistent trend of fewer ischemic events in this subgroup ( Figure 2 ).
| Variable | GPIs (n = 522) | No GPIs (n = 92) | p Value |
|---|---|---|---|
| Baseline variables | |||
| Age (yrs) | 59 (51–69) | 60 (53–72) | 0.64 |
| Age >75 yrs | 84 (16%) | 19 (21%) | 0.28 |
| Women | 96 (18%) | 24 (26%) | 0.09 |
| Weight (kg) | 76 (68–85) | 74 (65–85) | 0.32 |
| Systolic blood pressure (mm Hg) | 140 (120–159) | 134.5 (120–151) | 0.139 |
| Heart rate (beats/min) | 75 (65–86) | 78.5 (69–92) | 0.02 |
| Previous coronary artery bypass grafting | 2 (0.4%) | 2 (2%) | 0.11 |
| Previous chronic obstructive pulmonary disease | 10 (2%) | 6 (6%) | 0.02 |
| Diabetes | 71 (13%) | 19 (21%) | 0.08 |
| Dyslipidemia | 200 (38%) | 40 (43%) | 0.35 |
| Hypertension | 212 (41%) | 50 (54%) | 0.01 |
| Previous myocardial infarction | 38 (7%) | 6 (6%) | 0.79 |
| Previous peripheral arterial disease | 18 (3%) | 10 (11%) | 0.004 |
| Previous PCI | 44 (8%) | 7 (7%) | 0.79 |
| Previous stroke | 6 (1%) | 2 (2%) | 0.34 |
| Current smokers | 253 (48%) | 38 (41%) | 0.20 |
| Known renal failure | 9 (2%) | 5 (5%) | 0.04 |
| Aspirin | 514 (98%) | 89 (97%) | 0.22 |
| Angiotensin-converting enzyme inhibitors or angiotensin receptor blockers | 434 (83%) | 60 (65%) | <0.0001 |
| β blockers | 485 (93%) | 61 (66%) | <0.0001 |
| Insulin | 64 (12%) | 19 (21%) | 0.03 |
| Killip class II, III, or IV on admission | 34 (6%) | 10 (11%) | 0.14 |
| Procedural variables | |||
| Abciximab | 453 (87%) | 0 | |
| Eptifibatide | 62 (12%) | 0 | |
| Tirofiban | 8 (1%) | 0 | |
| Clopidogrel | 515 (99%) | 85 (92%) | 0.002 |
| Clopidogrel loading dose (>600 mg) | 176 (34%) | 26 (28%) | 0.3045 |
| Culprit coronary artery | 0.69 | ||
| Left anterior descending | 211 (44%) | 25 (46%) | |
| Left circumflex | 72 (15%) | 5 (9.%) | |
| Right | 191 (40%) | 24 (44%) | |
| Embolectomy/thrombectomy device | 253 (53%) | 19 (35%) | 0.01 |
| Stent implanted | 457 (95%) | 51 (94%) | 0.74 |
| Unfractionated heparin (as opposed to enoxaparin) | 78 (17%) | 13 (25%) | 0.14 |
| End Point | GPIs (n = 522) | No GPIs (n = 92) | Relative Risk (95% Confidence Interval) | p Value |
|---|---|---|---|---|
| Death, ∗ complication of myocardial infarction, procedure failure, or major bleeding (primary end point) | 152 (29%) | 28 (30%) | 1.01 (0.72–1.41) | 0.97 |
| Death, ∗ recurrent myocardial infarction or ACS, or urgent revascularization | 37 (7%) | 15 (16%) | 0.81 (0.39–1.68) | 0.57 |
| Death, ∗ recurrent myocardial infarction, or urgent revascularization | 24 (5%) | 11 (12%) | 1.54 (0.46–5.12) | 0.48 |
| Death ∗ or recurrent myocardial infarction | 18 (3%) | 11 (12%) | 1.16 (0.34–3.96) | 0.81 |
| Any death ∗ | 14 (3%) | 9 (10%) | 1.33 (0.31–5.75) | 0.70 |
| Any complication of myocardial infarction | 31 (6%) | 6 (6%) | 0.91 (0.4–2.1) | 0.83 |
| Any procedure failure | 120 (25%) | 17 (31%) | 0.79 (0.54–1.13) | 0.20 |
| Major bleeding | 19 (4%) | 3 (3%) | 4.42 (0.48–41.01) | 0.19 |
| Minor bleeding | 46 (9%) | 5 (6%) | 2.4 (0.79–7.29) | 0.12 |
| Major or minor bleeding | 60 (12%) | 7 (8%) | 2.7 (0.99–7.38) | 0.05 |
Net clinical benefit was also similar. Major and minor bleeding rates were also similar, but the combination of major and minor bleeding demonstrated a trend toward increased bleeding with GPI use.
Baseline and procedural characteristics for the patients treated with GPIs with the TRA compared with non-TRA patients are listed in Table 3 . There were a few clinical differences between the groups; for example, a greater fraction of non-TRA patients were in higher Killip classes. The crude rates of events suggested that the TRA with GPIs was associated with far fewer end points than a nontransradial approach with GPIs. However, after adjustment with propensity score analysis, no statistically differences emerged between the groups ( Table 4 , Figure 3 ). Predictably, major bleeding was reduced significantly with the TRA compared with nontransradial approaches. Minor bleeding rates were similar in the 2 groups. There were significantly fewer blood transfusions with the TRA with GPIs compared with nontransradial approaches.
| Variable | TRA (n = 522) | Nontransradial Approach (n = 207) | p Value |
|---|---|---|---|
| Baseline variables | |||
| Age (yrs) | 59 (51–69) | 60 (53–71) | 0.32 |
| Age >75 yrs | 84 (16%) | 38 (18%) | 0.46 |
| Women | 96 (18%) | 57 (27%) | 0.01 |
| Weight (kg) | 76 (68–85) | 75 (66–87) | 0.72 |
| Systolic blood pressure (mm Hg) | 140 (120–159) | 134.5 (120–155) | 0.10 |
| Heart rate (beats/min) | 75 (65–86) | 75 (65–87) | 0.96 |
| Previous coronary artery bypass grafting | 2 (0.4%) | 4 (2%) | 0.06 |
| Previous chronic obstructive pulmonary disease | 10 (1.9%) | 6 (2.9%) | 0.41 |
| Diabetes | 71 (13%) | 34 (16%) | 0.33 |
| Dyslipidemia | 200 (38%) | 90 (43%) | 0.20 |
| Hypertension | 212 (41%) | 103 (50%) | 0.02 |
| Previous myocardial infarction | 38 (7%) | 15 (7%) | 0.99 |
| Previous peripheral arterial disease | 18 (3%) | 6 (3%) | 0.71 |
| Previous PCI | 44 (8%) | 18 (9%) | 0.91 |
| Previous stroke | 6 (1%) | 7 (3%) | 0.06 |
| Current smokers | 253 (48%) | 95 (46%) | 0.53 |
| Known renal failure | 9 (2%) | 5 (2%) | 0.55 |
| Aspirin | 514 (98%) | 198 (96%) | 0.03 |
| Angiotensin-converting enzyme inhibitors or angiotensin receptor blockers | 434 (83%) | 153 (74%) | 0.01 |
| β blockers | 485 (93%) | 179 (86%) | 0.01 |
| Insulin | 64 (12%) | 36 (17%) | 0.07 |
| Killip class II, III, or IV | 34 (6%) | 37 (18%) | <0.0001 |
| Procedural variables | |||
| Abciximab | 453 (87%) | 165 (80%) | 0.02 |
| Eptifibatide | 62 (12%) | 40 (19%) | 0.01 |
| Tirofiban | 8 (1%) | 2 (1%) | 0.73 |
| Clopidogrel | 515 (98.7%) | 185 (89%) | <0.0001 |
| Clopidogrel loading dose (>600 mg) | 176 (34%) | 16 (8%) | <0.0001 |
| Culprit coronary artery | 0.01 | ||
| Left main | 5 (1%) | 6 (3%) | |
| Left anterior descending | 211 (44%) | 65 (34%) | |
| Circumflex | 72 (15%) | 21 (11%) | |
| Right | 191 (40%) | 95 (50%) | |
| Bypass graft | 0 (0%) | 2 (1%) | |
| Embolectomy/thrombectomy device | 253 (53%) | 72 (38%) | 0.01 |
| Stent implanted | 457 (95%) | 179 (95%) | 0.78 |
| Heparin | 78 (17%) | 30 (17%) | 0.93 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
