Samer S. Kabbani

Platelet Reactivity Characterized Prospectively A Determinant of Outcome 90 Days After Percutaneous Coronary Intervention

Background—Platelet activation is pivotal in the pathogenesis of complications after percutaneous coronary interventions (PCI). We previously reported substantial interindividual variability in activation of glycoprotein (GP) IIb/IIIa in response to a low concentration of ADP. We assessed GP IIb/IIIa activation prospectively to determine whether this could differentiate patients at low risk from those at high risk for complications early and late after PCI. Methods and Results—A total of 112 patients undergoing PCI were studied. Platelet reactivity was determined with the use of flow cytometry. Patients were classified into high and low platelet reactivity groups on the basis of extent of activation of GP IIb/IIIa in response to 0.2 &mgr;mol/L ADP. The median value was used for differentiation. The incidence during 90-day follow-up interval of a composite end point (myocardial infarction, urgent revascularization, or repeat revascularization) was determined in each group. Follow up was completed in all 112 patients. The 2 groups were similar with respect to diverse clinical characteristics. Nevertheless, the incidence of the composite end point occurred in 26.8% of the high and 7.1% in the low platelet reactivity group (P =0.01). The difference in the composite end point was most striking during the 30- to 90-day interval after PCI (16.7% versus 1.9%;P =0.02). Repeat revascularization was more frequent in those with increased platelet reactivity (17.9% versus with 3.6%, P =0.029). Conclusions—Prospective assessment of platelet GP IIb/IIIa activation permits stratification of patients into low- and high-risk groups with respect to adverse events after PCI.

Suboptimal early inhibition of platelets by treatment with Tirofiban and implications for coronary interventions

Inhibition of aggregation of platelets from 15 to 60 minutes after onset of treatment with tirofiban is less than that after onset of treatment with abciximab. The suboptimal inhibition occurs during a critical interval, immediately after iatrogenic vessel injury. Further, the suboptimal inhibition is likely to be responsible, at least in part, for the greater incidence of periprocedural myocardial infarction seen in the TARGET trial after treatment with tirofiban compared with abciximab. The difference observed is a reflection of the intrinsic pharmacologic characteristics of the 2 agents. Altering pharmacodynamics by dosage adjustment leading to higher concentrations in blood of tirofiban during the first hour after treatment onset requires further study.

Increased concentrations of tirofiban in blood and their correlation with inhibition of platelet aggregation after greater bolus doses of tirofiban

P in the do Tirofiban And Reopro Give similar Efficacy outcomes Trial (TARGET) treated with abciximab compared with tirofiban had a lower 30-day incidence of the combined end point of death, myocardial infarction, and target vessel revascularization that reflected, in large part, a decrease in the incidence of periprocedural myocardial infarction.1 We have found that the extent of platelet aggregation inhibition was greater from 15 to 60 minutes after abciximab than that after the TARGET study dosage of tirofiban (10 g/kg bolus, followed by a 0.15 g/ kg/min infusion).2–4 To identify a bolus associated with maximal inhibition of platelet aggregation, we used flow cytometry to characterize the inhibitory effects of tirofiban in blood from patients with acute coronary syndromes. In subsequent studies, the average extent of inhibition of turbidometric aggregation in response to 20 M of adenosine diphosphate (ADP) was found to be 90% during the first hour when a bolus of 25 g/kg was combined with the same infusion (0.15 g/kg/min for 18 to 24 hours) employed in the TARGET study.5 In this report, we describe the concentrations of tirofiban in blood and their correlation with the inhibition of turbidometric aggregation in response to 20 M of ADP. • • • Protocols were approved by the institutional review boards of the participating institutions. All patients provided written informed consent. Eligible patients included those with an acute coronary syndrome (ischemic symptoms plus either 0.5 mm of ST-segment depression on electrocardiogram or increased troponin or creatine kinase-MB) in whom a percutaneous coronary intervention was clinically mandated and performed. Exclusion criteria included treatment with antiplatelet agents other than aspirin in the previous 14 days, thrombolytic therapy within 24 hours, renal insufficiency (creatinine 2.5 mg/dl), and contraindication to treatment with a glycoprotein IIb/IIIa inhibitor. To perform in vitro studies, blood was obtained from patients with acute coronary syndromes before coronary intervention or treatment with an antiplatelet agent other than aspirin. After anticoagulation with corn trypsin inhibitor, a specific inhibitor of coagulation factor XIIa,6 the blood was spiked with 0, 50, 100, or 150 ng/ml of tirofiban. After incubation for 15 minutes at room temperature, the capacity to bind fibrinogen in response to 1 M of ADP was determined with the use of flow cytometry. The concentration of tirofiban in blood and its associated inhibition of platelet function were assessed in a preliminary study in which patients were treated with a 10 g/kg bolus2 and in a subsequent study in which patients were treated in a sequential, open-label, nonrandomized fashion with a 20 or 25 g/kg bolus.5 Tirofiban was administered before the start of the coronary intervention, and each bolus was combined with an 18to 24-hour infusion of 0.15 g/kg/min. Patients were treated with aspirin (325 mg before the procedure and daily) and unfractionated heparin (target activated clotting time 250 seconds). Clopidogrel (300 mg initially and then 75 mg daily) was begun after the coronary intervention ( 1 hour). The concentration of tirofiban was determined by radioimmunoassay after 5, 15, 30, 45, 60, 120, and 360 minutes.7 Platelet function was assessed by light transmission aggregometry (Bio Data Corp., Helena Laboratories, Horsham, Pennsylvania), and flow cytometry before treatment and after 15, 30, 45, 60, and 120 minutes, and an additional assessment was made after 5 minutes with flow cytometry. Blood for aggregometry was anticoagulated with D-Phe-Pro-Argchloromethyl ketone (PPACK, 38 M) to avoid effects of citrate on inhibitory properties of tirofiban.8 Blood for flow cytometry was anticoagulated with corn trypsin inhibitor. Maximal turbidometric ex vivo aggregation after 4 minutes was assessed in plateletrich plasma in response to 20 M of ADP (Chronolog, Havertown, Pennsylvania). Assessment of the capacity of platelets to bind fibrinogen was performed as previously described in detail.9 Assay and fixation were performed at each site for flow cytometry. Samples were analyzed at the University of Vermont within 30 hours. As previously described,5 the primary objective was to identify a bolus dose of tirofiban that achieved a mean inhibition of platelet aggregation of 90% plus a lower bound of the 95% confidence interval of inhibition of 85% from 15 to 45 minutes after onset of treatment. The slope of the association between selected concentrations of tirofiban and the extent of From the University of Vermont, Burlington, Vermont; University of Pennsylvania Medical Center, Philadelphia, Pennsylvania; Baylor College of Medicine, Houston, Texas; Prairie Cardiovascular, Springfield, Illinois; and Merck & Company, Inc., West Point, Pennsylvania. This report was supported by Merck & Company, Inc., West Point, Pennsylvania. Dr. Schneider’s address is: University of Vermont, 208 South Park Drive, Suite 2, Colchester, Vermont 05446. E-mail: djschnei@zoo.uvm.edu. Manuscript received August 7, 2002; revised manuscript received and accepted September 24, 2002.

Usefulness of platelet reactivity before percutaneous coronary intervention in determining cardiac risk one year later.

F low cytometric analysis of platelets with the use of epitope-dependent monoclonal antibodies or fluorochrome-labeled ligands provides sensitive and specific assessment of platelet activation.1,2 Assessment of platelet function has not yet been implemented to direct antiplatelet therapy. In contrast, analogous methods are used extensively in the diagnosis and classification of malignancies, as well as to guide antineoplastic therapy.3–5 We have previously reported that flow cytometric analysis of platelet reactivity prospectively identifies those patients at high and low risk of cardiac events during the first 90 days after percutaneous coronary intervention (PCI).6 The aim of the present report was to determine whether this method provides an outcome measure at 6 months and 1 year of follow-up in patients who underwent PCI. • • • The institutional review board of the University of Vermont approved the protocol. All patients provided informed consent. Eligible nonconsecutive patients included those with symptomatic coronary artery disease, a normal creatine kinase and creatine kinase-MB at the time of the procedure, and no pretreatment with intravenous glycoprotein (GP) IIb/IIIa inhibitors. Blood was obtained shortly before PCI for assay of platelet reactivity by flow cytometry. Patients were classified into 1 of 2 groups, high reactivity and low reactivity based on the percentage of platelets capable of binding fibrinogen in response to adenosine diphosphate 0.2 M (activation of platelet GP IIb/IIIa). The median value for all patients was used to stratify the subjects. PCI was performed as clinically mandated. All patients were treated with aspirin 325 mg/daily in addition to clopidogrel 75 mg/daily for 4 weeks. The composite of myocardial infarction (MI), urgent revascularization, and repeat vascularization was assessed after 6 months and 1 year. Follow-up was obtained by telephone calls and chart review at 3, 6, and 12 months. Follow-up was completed in all but 3 of the 112 patients studied ([97.3%] 2 deaths, 1 lost to follow-up). Platelet reactivity was assessed as previously described.6 In brief, blood was anticoagulated with corn trypsin inhibitor. Platelet reactivity was determined with respect to the activation of GP IIb/IIIa. Blood in 5l aliquots was added to microcentrifuge tubes containing HEPES-Tyrodes buffer and fluorochrome-labeled ligands. Adenosine diphosphate (0.2 and 1 M) was used to activate platelets. A peridinin chlorophyll protein conjugated antibody to GP IIIa was used as an activation-independent marker of platelets. Fluorescein isothiocyanate conjugated fibrinogen was used to assess the activation of GP IIb/IIIa. The reaction mixture was incubated at room temperature for 15 minutes. Subsequently, platelets were fixed and red blood cells were lysed with Optilyse-C (Immunotech, Marseille, France). Association of ligands with platelets was determined with a fluorescence-activated cell sorter. Platelets were identified on the basis of particle size and on association with CD61 antibody. Descriptive statistics were implemented for all measures. Comparison of the cumulative incidence of events in high and low reactivity groups was conducted using 2 2 contingency table methods and Fisher’s exact test for dichotomous measures. Comparison of the incidence of clinical outcomes over time in the high and low reactivity groups were From the University of Vermont College of Medicine, Burlington, Vermont. Dr. Kabbani’s address is: 111 Colchester Avenue, Cardiology Unit—Fletcher Allen Health Care—McClure 1, Burlington, Vermont 05401. E-mail: Samer.Kabbani@vtmednet.org. Manuscript received September 13, 2002; revised manuscript received and accepted December 4, 2002. FIGURE 1. Kaplan-Meier curves of the probability of freedom from the composite end point at 1 year. Platelet reactivity was assessed by flow cytometric determination of the capacity of platelets to bind fibrinogen in response to adenosine diphosphate 0.2 M. The divergence of the curves during the initial 24 hours predominantly reflects a higher incidence of periprocedural MI in those with high platelet reactivity. The highest incidence of cardiac events occurred during the first 6 months after PCI.

Platelet reactivity in Coronary ostial blood: A reflection of the thrombotic state accompanying plaque rupture and of the adequacy of anti-thrombotic therapy

Background: Optimal anti-thrombotic therapy for acute coronary syndromes (ACS) should suppress pro-thrombotic activity at the site of plaque rupture. We sought to determine whether platelet reactivity is increased in blood in the immediate vicinity of a ruptured plaque and is apparent even when blood is obtained by sampling from a catheter placed proximal to the lesion. Methods: Blood was obtained from a catheter placed in the aorta and from the same catheter after engaging the culprit coronary artery. Platelet reactivity was determined with the use of flow cytometry by surface expression of P-selectin. Results: In preliminary studies we demonstrated that a marker of thrombin activity, fibrinopeptide A, was similarly increased in blood taken from the coronary sinus and coronary arterial ostium of patients with ACS. Subsequently blood was obtained from the aorta and coronary arterial ostium through a coronary guide catheter for assessment of platelet reactivity in 23 subjects with ACS and 22 subjects with stable angina. The percentage of platelets expressing P-selectin in response to 0.2[emsp4 ]μM adenosine diphosphate (ADP) was greater in coronary arterial samples from patients with ACS (aorta=6.1±1%, coronary artery=8.8±1.6%, p=0.02) compared with that in patients with stable symptoms (aorta=6.9±1.2, coronary artery=6.5±1.4, p=NS). Conclusions: Coronary arterial blood obtained from the ostium through a coronary guide catheter can be used to determine whether thrombin activity and platelet reactivity are increased in the immediate vicinity of a ruptured atherosclerotic plaque. The simplicity of the approach developed should facilitate its use in future studies designed to determine the impact of optimal suppression of platelet reactivity and the pro-thrombotic state before coronary interventions on short- and long-term clinical outcomes.AbstractBackground: Optimal anti-thrombotic therapy for acute coronary syndromes (ACS) should suppress pro-thrombotic activity at the site of plaque rupture. We sought to determine whether platelet reactivity is increased in blood in the immediate vicinity of a ruptured plaque and is apparent even when blood is obtained by sampling from a catheter placed proximal to the lesion. Methods: Blood was obtained from a catheter placed in the aorta and from the same catheter after engaging the culprit coronary artery. Platelet reactivity was determined with the use of flow cytometry by surface expression of P-selectin. Results: In preliminary studies we demonstrated that a marker of thrombin activity, fibrinopeptide A, was similarly increased in blood taken from the coronary sinus and coronary arterial ostium of patients with ACS. Subsequently blood was obtained from the aorta and coronary arterial ostium through a coronary guide catheter for assessment of platelet reactivity in 23 subjects with ACS and 22 subjects with stable angina. The percentage of platelets expressing P-selectin in response to 0.2[emsp4 ]μM adenosine diphosphate (ADP) was greater in coronary arterial samples from patients with ACS (aorta=6.1±1%, coronary artery=8.8±1.6%, p=0.02) compared with that in patients with stable symptoms (aorta=6.9±1.2, coronary artery=6.5±1.4, p=NS). Conclusions: Coronary arterial blood obtained from the ostium through a coronary guide catheter can be used to determine whether thrombin activity and platelet reactivity are increased in the immediate vicinity of a ruptured atherosclerotic plaque. The simplicity of the approach developed should facilitate its use in future studies designed to determine the impact of optimal suppression of platelet reactivity and the pro-thrombotic state before coronary interventions on short- and long-term clinical outcomes.

Abciximab-Associated Pseudothrombocytopenia

A60-year-old woman was transferred for coronary angiography after a non–Q-wave myocardial infarction. She was treated with aspirin, metoprolol, unfractionated heparin, and abciximab administered as a weight-based bolus followed by infusion. A complete blood count in blood anticoagulated with EDTA (purple-top Vacutainer) was obtained shortly after her arrival in the catheterization laboratory, 4 hours after the abciximab bolus. The reported platelet count had decreased from 440 000/mL to 21 …

Quantification by flow cytometry of the efficacy of and interindividual variation of platelet inhibition induced by treatment with tirofiban and abciximab.

BackgroundAfter exposure of platelets to abciximab and tirofiban in vitro, we have observed variable inhibition of fibrinogen binding and a lack of inhibition of α‐granule degranulation. DesignTo determine whether such changes occur with treatment, platelet reactivity was assayed in blood from 50 patients receiving abciximab or tirofiban. MethodsPlatelet reactivity was determined before and during steady‐state infusions of abciximab (0.125 μg/kg/min) or tirofiban, with either the PRISM‐PLUS dosage (0.1 μg/kg/min) or the RESTORE dosage (0.15 μg/kg/min). Fibrinogen binding and P‐selectin expression were determined by flow cytometry after stimulation of platelets with ADP (0.2 or 1 μM) or thrombin‐receptor agonist peptide (TRAP, 25 μM). ResultsBoth dosages of tirofiban and abciximab reduced fibrinogen binding in response to 0.2 μM ADP comparably. However, fibrinogen binding in response to 1.0 μM ADP or 25 μM TRAP was inhibited to a greater extent by the RESTORE dosage of tirofiban and abciximab than by the PRISM‐PLUS dosage of tirofiban (P  < 0.05). Furthermore, only the RESTORE dosage of tirofiban and abciximab reduced P‐selectin expression in response to ADP. Inhibition with each regimen varied markedly between patients. ConclusionsThe RESTORE dosages of tirofiban and abciximab each inhibit fibrinogen binding and α‐granule degranulation similarly. However, substantial interindividual variation in inhibition of fibrinogen binding is evident.