PCI with and without Abciximab after Upstream Eptifibatide Use: Outcomes in High-Risk Patients
- Volume 18 - Issue 12 - December, 2006
- Posted on: 8/1/08
- 1 Comments
- 5751 reads
There was a significantly higher percentage of patients with acute or MI (within 7 days before the index PCI) in the EA group compared to the EE group: 80% versus 53%; p < 0.001. Meaningful differences in procedural characteristics between the two treatment groups included a greater degree of lesion complexity, including a greater number of ACC Type-B2 and C lesions, and lesions containing thrombus in the EA group compared to the EE group. The clinical and procedural characteristics of the two groups stratified by clinical presentation are shown in Table 1. Significant differences in baseline patient and procedural characteristics between the two treatment groups were present in the unstable angina patients, reflecting the strategy of use of abciximab in patients with higher-risk lesion and procedural characteristics. Baseline characteristics in the subgroup of patients with MI were reasonably similar between the two treatment groups.
For the entire study population, the change in overall platelet count from baseline to after PCI was not statistically different for either treatment group. The average decrease in hemoglobin postprocedure was 1.0 g/dl in both treatment groups; p = 0.284. TIMI bleeding rates were similar for the two treatment groups: 4.1% EA versus 2.6% EE; p = 0.25. The incidence of any vascular complications was comparable for the two treatment groups. No adverse events occurred in-lab for either of the two treatment groups. The in-hospital outcomes for patients stratified by clinical presentation are shown in Table 2. In-hospital outcomes for both treatment groups in patients with unstable angina paralleled the overall outcomes outlined above and were not significantly different. For the MI subgroup, overall MACE was lower for EA, 0.9%, than for EE, 4.3%; p = 0.017.
For the entire study population, overall MACE at 30 days was 4.8% for the EE group and 4.2% for the EA group (p = 0.66), and at 1 year, overall MACE rates were similar (25.2% for the EE group and 26.0% for the EA group; p = 0.82). The early and late clinical outcomes following PCI stratified by clinical presentation are shown in Table 3. Kaplan-Meier plots of MACE-free, MI or death-free and revascularization-free survival for patients stratified by clinical presentation and treatment group are shown in Figure 2. In patients with unstable angina, overall MACE at 30 days was lower for the EE group (3.4%) than for the EA group (10.0%) because of a significantly higher rate of nontarget revascularization in the EA group compared to the EE group. At 1 year, overall MACE remained lower in the EE group than the EA group (23.9% vs. 41.7%, respectively; p = 0.013), primarily because of a lower rate of revascularization during follow up in the EE group than in the EA group (20.3% vs. 33.3%, respectively; p = 0.054). In patients with MI, overall MACE was similar at 30 days in the two treatment groups, although recurrent MI or death was lower in the EA group than in the EE group (0.9% vs. 4.4%, respectively; p = 0.017). At 1 year, overall MACE was nonsignificantly lower in the EA group than in the EE group (21.7% vs. 26.3%, respectively; p = 0.285). At 1 year, there was a significantly lower incidence of recurrent MI or death in the EA group than in the EE group (6.9% vs. 15.6%, respectively; p = 0.007).
Because of the difference in baseline characteristics in the two treatment groups, we also assessed risk-adjusted outcomes. Standard multivariate analysis using Cox proportional hazards modeling of any MACE at 1 year identified multivessel disease (two- or three-vessel disease) and diabetes as independently predictive of increased overall MACE, and stent use and current smoking as independently predictive of decreased overall MACE (Table 4). A history of MI, diabetes and a MI within 7 days of the index PCI were all independently predictive of a recurrent MI or death. Stent use was independently predictive of a decrease in any revascularization: hazard ratio (HR) 0.57 (0.40–0.80).
The risk-adjusted HRs of outcomes at 1 year, EA compared to EE, stratified by clinical presentation, are shown in Figure 2. In patients with unstable angina, the risk-adjusted hazard of overall MACE at 1 year was 1.80 (1.09–2.96), EA compared to EE. The risk-adjusted hazard of any revascularization, however, was no longer significantly higher in the EA group than the EE group: 1.64 (0.95–2.84). In patients with an MI within 7 days of their index PCI, the risk-adjusted HR of overall MACE was 1.16 (0.86–1.56), EA compared to EE. The risk-adjusted HR of MI or death at 1 year remained significantly lower in the EA group compared to the EE group, HR 0.53 (0.28–0.98).
Receptor-binding studies. The receptor-binding study results are shown in Table 5 and Figure 3. In Figure 3, the number of receptor-binding sites identified as MAB1 and MAB2 are shown prior to abciximab administration and 10 minutes and 8 hours after abciximab infusion, with and without pretreatment with eptifibatide. There were no differences in MAB2 at any time period after abciximab use in the absence of eptifibatide. There was a significant increase in MAB2 after abciximab administration following pretreatment with eptifibatide, consistent with washout of eptifibatide.
Receptor occupancy increased from 11.2 ± 6.7% to 94.7 ± 1.6% after abciximab without eptifibatide at 10 minutes (p < 0.05), and fell slightly but significantly to 88.0 ± 5.4% at 8 hours (p < 0.05 vs. baseline) (Table 5). Receptor occupancy at baseline following eptifibatide infusion, with an average duration of 31 ± 10 hours, was 67.2 ± 8.5%. After abciximab, receptor occupancy rose to 93.2 ± 1.8% at 10 minutes (p < 0.05 vs. baseline eptifibatide), and again fell slightly, but significantly, to 85.1 ± 5.9% at 8 hours (p < 0.05 vs. baseline and 10 minutes). Receptor occupancy after abciximab with eptifibatide pretreatment was not different than with abciximab alone at 10 minutes or at 8 hours.
Ultegra RPFA. PAU values for the two treatment strategies are shown in Table 5. After abciximab alone, PAU fell from 195 ± 47 units to 23 ± 15 units (p < 0.05) at 10 minutes, and remained low at 8 hours (PAU 31 ± 22; p = NS) versus 10 minutes. Baseline PAU values were available in only 6 patients in the EA group, but the absolute PAU values for all 25 patients at 10 minutes (16 ± 17) and 8 hours (20 ± 17) were similar to the PAU values seen in the abciximab-alone group at the same time periods (p = NS).