The Relative Effects of Abciximab and Tirofiban on Platelet Inhibition and C-Reactive Protein during Coronary Intervention
- Volume 22 - Issue 1 - January, 2010
- Posted on: 1/7/10
- 0 Comments
- 10698 reads
ABSTRACT: Background. We sought to compare the efficacy of tirofiban and abciximab on platelet inhibition as well as their effects of platelet inhibition on C-reactive protein levels during percutaneous coronary intervention (PCI). Methods. Using a randomized, double-blind study design, 95 consecutively eligible patients were randomized to receive either tirofiban or abciximab before undergoing native coronary artery revascularization with a stent. Clinical endpoints were death, nonfatal MI, target vessel revascularization (TVR) with coronary artery bypass grafting or PCI within 30 days of the study procedure. The medications were compared for differences in platelet aggregation as measured by a rapid function platelet assay, as well as measurements of the inflammatory marker C-reactive protein (CRP) at frequent intervals following drug administration during PCI. Results. A total of 95 patients were randomized to abciximab (n = 44) or tirofiban (n= 51). There was no significant difference in platelet aggregation documented throughout the procedure (10-, 20-, 30-, 45-minute time points). In diabetic patients abciximab had significantly lower platelet inhibition as compared to tirofiban at 10 minutes (84.17 ± 8.28% vs. 90.40 ± 5.79%; p = 0.0097). Using a Spearman correlation coefficient model, hs-CRP demonstrated an inverse relationship with platelet inhibition over time (-0.7307, p=0.0002) in patients treated with abciximab. Conclusion. There is no major difference in platelet inhibition between tirofiban and abciximab during PCI. In this study, tirofiban showed a greater inhibition in diabetic subsets at the first time point within PCI. Platelet inhibition may be inversely related to the levels of CRP during PCI.
J INVASIVE CARDIOL 2010;22:2–6
Key words: glycoprotein IIb/IIIa receptor antagonists,
platelet inhibition, inflammation, percutaneous coronary intervention
Several randomized trials have studied the use of glycoprotein (GP) IIb/IIIa platelet inhibitors in unstable angina/non-Q-wave myocardial infarction (MI) and in the cardiac catheterization laboratory.1–9 Results from a multicenter study demonstrated that a high degree of platelet inhibition during percutaneous coronary intervention (PCI) was associated with less major adverse cardiac events (MACE).10 Although all GP IIb/IIIa inhibitors have been reported to produce > 80% inhibition of platelet aggregation ex vivo and in vitro, the in vivo degree of GP IIb/IIIa inhibition and platelet aggregation during the PCI itself have not been well studied.
The aim of this study was to compare the efficacy of abciximab (a noncompetitive inhibitor of fibrinogen binding) (Reo-Pro, Eli Lilly and Company, Indianapolis, Indiana) and tirofiban (a competitive inhibitor of fibrinogen binding) (Aggrastat, Merck, White House Station, New Jersey), both in combination with heparin, in inhibiting platelet aggregation and inflammatory marker fluctuation during the duration of the PCI procedure in patients undergoing urgent PCI for unstable angina as well as elective PCI.
ATHENS (abciximab versus tirofiban for inhibition of platelets during native coronary interventions) is a pilot study with a randomized, double-blind design. Patients were considered eligible for the study if they were undergoing an elective PCI with a stent, or had PCI for unstable angina with normal baseline levels of CK-MB. Patients were excluded from the study if they had a primary lesion not amenable to PCI/stent implantation, prior history of thrombolytic therapy within 3 days, therapy with abciximab within the previous 15 days, therapy with eptifibatide or tirofiban within the previous 24 hours, elevated CK-MB at the time of the procedure, serum creatinine > 2.0 mg/dl, or any contraindication to anticoagulation. Diabetic patients were a prespecified subgroup for our study.
Using a double blind 3:2 randomization, patients who met the eligibility criteria were randomized between tirofiban and abciximab and were continued on blinded treatment through 12 hours. Randomization was performed by the research pharmacy based on a list of random assignments provided by the data center. The concentration of the study drugs were adjusted so that the bolus and infusion volumes of the two regimens would be identical for a given body weight. All labels indicated “study drug” and were labeled with the calculated infusion rate by the research pharmacy.
Anticoagulation regimen. The study medications were administered as a bolus plus infusion for 12 hours. Heparin therapy was also administered in the beginning of the PCI procedure to achieve a target activated clotting time (ACT) of 200–250 (sec/i). Abciximab was dosed as 0.25 mg/kg bolus given immediately before the PCI, followed by 0.125 µg/kg/min (max 10 mg) for 12 hours. A 10µg/kg bolus of tirofiban was given as a bolus, followed by a 0.15µg/kg/min infusion for 12 hours.9 PCI commenced immediately following bolus administration of the GP IIb/IIIa antagonist.
Platelet aggregation. Blood samples were obtained at up to 6 time points: baseline (prior to study drug bolus), 10, 20, 30, and 45 minutes later, as well as at the end of the procedure if the procedure lasted longer then 45 minutes. Using methods previously described and validated, samples were tested with a rapid function platelet assay, which evaluates the fibrinogen-induced platelet aggregation (Ultegra, Accumetrics, San Diego, California). Briefly, pharmacological blockade of GP IIb/IIIa receptors prevents the interaction between platelet GP IIb/IIIa receptors and fibrinogen-coated beads thereby diminishing agglutination in proportion to the degree of receptor blockade achieved.11 The light absorbance of the sample is measured as a function of time, and the rate of agglutination is quantified as platelet activation units (PAUs). Platelet inhibition can be calculated as a percentage of the baseline value.
Differences in platelet aggregation were measured by a rapid function platelet assay at prespecified time points (10, 20, 30, 45 minutes, end of procedure), and measurement of the inflammatory marker high-sensitivity C-reactive protein (hs-CRP) at the same prespecified time points listed above. Platelet aggregation was determined at the time the blood sample was drawn in the same facility as the catheterization laboratory. Based on previous trials suggesting a difference in platelet aggregation between diabetics and nondiabetics receiving GP IIb/IIIa inhibitors, differences between the study medications on platelet aggregation were analyzed in this subgroup.
Endpoints. Clinical endpoints included death, nonfatal myocardial infarction, target vessel revascularization (TVR) with coronary artery bypass grafting (CABG), or PCI within 30 days of the index procedure. A new myocardial infarction was defined as a new Q-wave on electrocardiogram (ECG) or a CK-MB rise > 5 times the upper limit of normal (ULN) in the presence of ischemic symptoms.
Inflammatory marker determination. Briefly, 5 cc of blood was collected in PPACK-containing tubes, spun to 3,000 rpm for 20 min and shock frozen to -70º C until inflammatory marker determination; hs-CRP was analyzed on the Dade-Behring BN 100 Nephelometer, using kits purchased from Dade-Behring. The assay was carried out per the manufacturers’ instructions and the laboratory was blinded to any clinical information.
Statistical analysis. For p-value calculations of the primary outcomes (MACE) a Fisher’s exact test of proportions was used, and unpaired Student’s t-test was used for continuous variables. To correlate the level of platelet inhibition and inflammatory markers a generalized estimation equation model was used. The dependent variable was platelet inhibition and the independent variable we aimed to model was the biomarker hs-CRP. Values were correlated with platelet inhibition at different time points. We used an unstructured working correlation in the model estimation. We also used Spearman rank-order correlation as a nonparametric measure of association based on the ranks of the data values.
A total of 95 patients were enrolled over 12 months. Of the patients enrolled, 44 (46%) were randomized to abciximab, 51 (53%) to tirofiban. The baseline clinical characteristics were similar between the two groups (Table 1). All procedures were successfully completed and all stents were placed without loss of side branches, slow-flow, or no-reflow. Study medications were given as assigned to those enrolled in the trial and every patient completed a full course of their designated therapy.
Clinical outcomes. Adverse clinical events were infrequent. There was one major in hospital adverse cardiac event (non-Q-wave MI with CK-MB > 5 x normal), in the tirofiban group (2%) (p = 0.68). CK-MB levels 2 times the ULN occurred in 2 (4.5%) patients in the abciximab and 5 (9.8%) patients in the tirofiban group (p = 0.44). There was 1 (2.3%) minor bleeding complication and 1 (2.3%) vascular complication in the abciximab group, and 2 (3.9%) vascular complications in the tirofiban group. In order to address the issue of different definitions of periprocedural myonecrosis, we also used a definition of 3 times the ULN of CK-MB; this adjustment did not qualitatively alter the results of the present study.
With respect to 30-day follow up, out-of-hospital events were infrequent; no deaths or MI occurred. There was one (2%) target vessel revascularization with PCI in the tirofiban group. There were 3 (6.8%) non-TVR with PCI in the abciximab group and 2 (3.9%) in the tirofiban group (p = 0.66).
Platelet inhibition. In the analysis of the entire patient population, platelet aggregation (PA) was similar between abciximab and tirofiban at all time points. Individual variability of platelet inhibition was documented at every time point, but no significant differences were shown between the abciximab and tirofiban groups (Figures 1 and 2). Mean platelet inhibition values for abciximab at 10, 20, 30, 45 minutes, and end of procedure were 86.75 ± 8.36, 87.69 ± 7.86, 87.42 ± 7.71, 84.86 ± 9.36, and 85.73 ± 8.09, respectively; for tirofiban these values were 88.46 ± 6.59, 87.33 ± 5.85, 85.36 ± 6.01, 89.54 ± 6.21, 86.17 ± 7.34, respectively.
In a subset analysis, diabetic patients had similar average platelet inhibition as nondiabetic patients except for the 45 minute time point that had significantly lower platelet inhibition documented in diabetics (83.8 ± 9.51% vs. 90.7 ± 4.47%; p = 0.047). Within the diabetic patient subset, the abciximab group had significantly lower platelet inhibition as compared to tirofiban at 10 minutes (84.17 ± 8.28% vs. 90.40 ± 5.79%; p = 0.0097) and a trend towards lower platelet inhibition at every other time point (Figures 3 and 4).
Inflammatory marker. Patients receiving abciximab and tirofiban had similar measurements of the inflammatory marker CRP at each time point (Table 2). In general, in-vivo measurement of CRP was associated with wide variability.
Within the abciximab group, greater platelet inhibition over time was correlated with decreasing CRP values. Using a Generalized Estimation Equation model, CRP demonstrated an inverse relationship with platelet inhibition over time (-0.7307, standard error 0.196; p = 0.0002). An inverse correlation was not observed with tirofiban use and CRP.
The present study demonstrated no overall difference between abciximab and tirofiban in platelet inhibition. However, there was a great deal of individual variability in both groups at various time points during PCI, with several patients having platelet inhibition values below the 80% and 90% level. Whether this variability would translate to procedural or 30-day clinical events was beyond the scope of this study.