Disaggregation of In Vitro Preformed Platelet-Rich Clots by Abciximab Increases Fibrin Exposure and Promotes Fibrinolysis
Abstract—The glycoprotein IIb/IIIa receptor inhibitor abciximab has been shown to facilitate the rate and the extent of pharmacological thrombolysis with recombinant tissue plasminogen activator (rtPA) in patients with acute myocardial infarction. However, the underlying mechanisms remain not fully determined. We sought to demonstrate that this facilitating effect of abciximab could be related to its potential to modify the clot architecture and the clot physical properties. Compared with fibrin-rich clots, platelets dramatically modified the in vitro properties of the fibrin network, leading to a significant increase of the permeability (Ks) and the viscoelasticity (G′) indexes but also leading to the appearance of platelet aggregates (surface area [S.ag]). These modifications resulted in a 2.6-fold decrease of the fibrinolysis rate when rtPA (1 nmol/L) was added before the initiation of clotting. Adding aspirin (100 μg/mL) or abciximab (0.068 μmol/L) before the clotting of platelet-rich clots (PRCs) lowered Ks by 50% and 70%, respectively (P<0.01), G′ by 41% and 66%, respectively (P<0.01), and S.ag by 32% and 61%, respectively (P<0.01). As a consequence, the lysis speed was increased by 21% with aspirin (P<0.01) and 45% with abciximab (P<0.01). However, unlike aspirin, permeation of preformed PRCs with abciximab (0.068 μmol/L) decreased G′ (37%, P<0.01), Ks (35%, P<0.001) and S.ag (25%, P=NS) and resulted in a 27% (P<0.01) increase of the lysis speed when abciximab and rtPA (0.2 μmol/L) were simultaneously permeated. This effect was found to be time dependent and was observed only with early permeation, starting within the first 10 minutes of clotting. These changes in the physical properties of the PRC architecture suggest that fibrin is removed from the platelet-fibrin aggregates and reexposed into the surrounding fibrin network, increasing rtPA access to fibrin and therefore the fibrinolysis rate. The superiority of abciximab over aspirin in accelerating fibrinolysis of forming and preformed PRCs is related to its ability to modulate the interactions of fibrinogen and fibrin with platelets. These findings provide new mechanistic information on reperfusion therapy.
Presented in part at the 71st Scientific Sessions of the American Heart Association, Dallas, Tex, November 8–11, 1998.
- Received May 12, 2000.
- Accepted July 12, 2000.
Direct inhibition of platelet interactions with fibrin(ogen) by abciximab (Reopro) can restore coronary flow,1 as shown in randomized trials combining abciximab with either primary angioplasty or thrombolysis in acute myocardial infarction.2 3 4 5 However, the comprehensive mechanisms remain to be fully understood. In the present study, we sought to demonstrate that this effect of abciximab is related to its potential to modify platelet-fibrin(ogen) interactions, leading to changes of clot architecture and clot physical properties and facilitating pharmacological thrombolysis.
For the first step in our experiments, differences between the physical properties of fibrin-rich clots (FRCs) and platelet-rich clots (PRC) were studied to delineate the mechanical impact of platelets on the fibrin network properties and their consequences for the fibrinolysis rate. The second step in our experiments evaluated the consequences of the inhibition of fibrinogen interactions with platelets on the fibrin network properties by adding aspirin or abciximab before clotting. In the third set of experiments, preestablished PRCs were permeated with aspirin or abciximab to evaluate further their ability to modulate the interaction between platelets and fibrin. The impact of these pharmacological changes on the fibrinolysis rate was assessed at each step by either loading recombinant tissue plasminogen activator (rtPA) before the initiation of clotting or by permeating preestablished FRCs and PRCs with rtPA.
Abciximab (Reopro) was supplied by Eli Lilly as a 2 mg/mL solution and was dissolved in buffer 1 (0.15 mol/L NaCl and 0.01 mol/L Tris-HCl, pH 7.4). Aspirin (Aspegic) was supplied by Synthelabo (Meudon-La-Forêt) as a 100 mg/mL solution and was dissolved in buffer 1. Murine monoclonal antibodies (anti-CD9 human platelet receptor [Syb]) were provided by Dr Boucheix (CNRS, Villejuif, France). Goat monoclonal anti-mouse antibodies from Amersham Pharmacia were available as 1 mg/mL solution in buffer 1. Human thrombin (Enzyme Research Laboratories Inc) was stored at 1000 IU/mL. rtPA was purchased from Boehringer-Ingelheim. FITC and streptavidin-tricolor were from Sigma, and CaCl2 (0.456 mol/L solution) was from Assistance Publique-Hôpitaux de Paris.
Preparation of Fibrin-Rich and Platelet-Rich Thrombi
Blood was from 8 healthy informed volunteers who had no known bleeding disorder and who had not ingested antiplatelet medication for at least 10 days. It was anticoagulated with trisodium citrate (1 vol of 0.13 mol/L citrate for 9 vol blood) and centrifuged at 800g for 15 minutes to provide platelet-rich plasma (PRP). Recentrifugation of PRP at 10 000g for 15 additional minutes provided platelet-poor plasma (PPP), which was then filtered (0.22-μm filters). PPP was then mixed with the PRP of the same donor to adjust for the final platelet count. Addition of CaCl2 (20 mmol/L) and thrombin (0.125 IU/mL) to 0.12 mL of either PPP or PRP led to the formation of FRCs and PRCs.
Mechanical Properties of Clots
The permeability index (Ks) of plasma FRCs and PRCs was measured by the permeation technique.6 Briefly, clots were formed in thin glass microchambers (250 μm) and were permeated with buffer 1 at different gradients of pressure (Figure⇓ I, available online at http://atvb.ahajournals.org). The calculated Ks index (in centimeters squared) provides information on the fibrin network architecture (shape and size of the pores) and represents the surface of the gel allowing flow.6 The thrombus permeability index equation is as follows: Ks=(Q · L · η)/(A · ΔP · t), where Q is the volume of liquid (in milliliters) having the viscosity η (10−2 poise), flowing through the fibrin gel with length L (2.2 cm) and cross section A (0.03 cm2) in a given time t (in seconds) under a differential pressure ΔP (ranging from 4000 to 10 000 dyne/cm2).
FRCs and PRCs were formed between two 12-mm-diameter glass coverslips in the torsion pendulum device shown in Figure⇑ II (available online at http://atvb.ahajournals.org) by using the same clotting conditions as for the permeation experiments.7 Clots had a constant width of 1 mm. After the initiation of clotting, a momentary impulse was carefully applied to the torsion pendulum arm (air pressure), causing free oscillations of this arm with strains <3%. The frequency of these free oscillations and the rate at which they are damped are functions of the elastic and viscous properties of the clots and are independent of the amplitude of the initial displacement of the arm.7 The rigidity index (G′, in dynes per centimeter squared), which reflects the viscoelastic properties of the clot, was calculated from the recordings of these oscillations on a chart recorder.
Two different sets of experiment were conducted: (1) FRCs and PRCs that were formed with buffer or antiplatelet agents were processed for Ks and G′ measurements. (2) Preformed FRCs and PRCs were carefully permeated with buffer or antiplatelet agents at a constant pressure gradient of 3000 dyne/cm2 before being processed for Ks and G′ measurements. A small perfusion chamber was added to the torsion pendulum to allow perfusion of the clots at a constant gradient of pressure while the viscoelastic properties were constantly monitored (Figure⇑ II).
Clot Morphological Properties
Confocal Scanning Laser Microscopy Experiments
Clots used for permeation experiments were then permeated with a 2 mmol/L FITC solution dissolved in 0.01 mol/L Tris, 0.1 mol/L NaCl, and 1 mmol/L EDTA (pH 8) buffer for 30 minutes. The excess of dye was eliminated by extensive washing with buffer 1.8 FITC binds to fibrin through nonspecific interactions with the COOH residues. Labeled specimens were scanned with an ACAS 570 interactive laser cytometer (Meridian Instruments) equipped with confocal optics. Twenty-five optical sections were collected at intervals of 1.0 μm in the z-axis and were combined into 1 image, generating a 3D reconstructed image of the fibrin network.6
Platelet Localization Within the Fibrin Network
Platelets within the fibrin matrix were specifically localized by using the anti-CD9 antibodies coupled to biotin (25 μg/mL) that were incubated for 5 minutes at 37°C with the PRP before the initiation of clotting. After clotting occurred, anti-CD9 antibodies were revealed by permeation of the clot with streptavidin-tricolor; fibrin was labeled with FITC as described above. After washing, platelet-rich areas containing fibrin-FITC and tricolor platelets appeared yellow; fibrin-FITC areas without platelets appeared green.
Quantitative analysis of the average area of platelet aggregates (S.ag, in micrometers squared) was performed with the Visilog software (version 5.01, Noesis). Because aggregates and the surrounding fibrin agglomerated within the reconstructed images from confocal scanning laser microscopy, an optical algorithm based on “the watershed line” transformation technique was required to separate them before being measured.9 10
Fibrinolysis was assessed by monitoring of the viscoelastic index G′ every 4 minutes, and the fibrinolysis speed corresponded to the time taken for G′ to decrease to 50% of its maximal value.11 Clots were formed in the exact same conditions as for viscoelastic measurements, and 2 different assays were developed: (1) In the static lysis assay, rtPA was added with buffer or with antiplatelet agents before the initiation of clotting. rtPA concentration (1 or 2 nmol/L) was adjusted so that lysis started after the first 10 minutes of initiation of clotting.11 (2) In the dynamic lysis assay, preformed PRC was carefully permeated with rtPA and buffer or antiplatelet agents at a final concentration of 0.2 μmol/L, which is the in vivo predicted concentration from computer simulation of thrombolysis.12 The different rtPA concentrations used in these 2 assays reflect the difference of their design: in the static assay, rtPA is already in the fibrin fibers, whereas in the dynamic lysis assay, rtPA needs to be delivered to the fibrin network under pressure-driven permeation.13
Comparison of the Efficacy of Aspirin and Abciximab in Remodeling the Anatomy of the Clot
To simulate in vitro the dynamic changes in the structure and the lysis speed of an occlusive coronary thrombosis after it is exposed to antiplatelet agents in vivo, occlusive and fully hydrated FRCs and PRCs were used to evaluate the potential of aspirin and abciximab to either prevent fibrinogen interactions with platelets or to remove platelets from fibrin. Aspirin and abciximab were either added before clotting initiation or carefully infused through preformed clots. Permeation started 5 and 40 minutes after clotting initiation for a 20-minute time period at a constant gradient of pressure of 3000 dyne/cm2. Fibrinolysis was assessed by loading rtPA in combination with either aspirin or abciximab before the initiation of clotting (static lysis assay) or by permeating rtPA (0.2 μmol/L) in combination with aspirin or abciximab (dynamic lysis assay). Control clots were formed and/or permeated with buffer 1. G′ and Ks were measured before, during, and after permeation. Clots used for permeation experiments (without rtPA) were then processed for microscopy.
Aspirin concentrations (50 and 100 μg/mL) correspond to an intravenous administration of 500 and 1000 mg aspirin, which completely inhibits platelet cyclooxygenase activity.14 Abciximab concentrations (0.034 and 0.068 μmol/L) correspond to an intravenous bolus administration of 0.15 and 0.30 mg/kg, which produces near-binding saturation of platelet surface glycoprotein (GP) IIb/IIIa receptors and inhibits thrombosis in vivo.15 This was further ascertained by the use of the rapid platelet functional assay (Accumetrics), which confirmed that 99% of the platelet GP IIb/IIIa receptors were occupied with both abciximab concentrations, whereas 95% and 100% of the platelet receptors remained free with aspirin at 50 and 100 μg/mL, respectively.16 These abciximab concentrations were also shown to inhibit 75% and 100% of the mechanical effect of platelets on fibrin, respectively.17
All experiments were conducted with a range of platelets of 75 000 to 150 000/μL. A dose-related effect of platelets on mechanical properties and on fibrinolysis speed was found. Results with a final platelet count of 75 000/μL are presented here, especially because measurements of permeability constant Ks and morphological index S.ag were more accurate than with 150 000 platelets per microliter.
Statistical analysis was performed with StatView software. Continuous variables were expressed as mean±SEM, and group differences were determined by ANOVA. The Student t test for paired samples was used to take into account the heterogeneity among the plasma of the 8 donors. A value of P<0.05 was considered to indicate statistical significance.
Comparison Between FRC and PRC Properties
Intra-assay and interindividual variabilities in permeability and viscoelastic experiments were similar to those previously reported for plasma FRCs.8 The coefficients of variation for Ks and G′ were 12.5% and 9.2%, respectively, for a double measurement on a single sample (intra-assay variability, n=8). Measurement of Ks was performed in different chambers. The coefficient of variation of Ks was slightly higher when platelets were added before clotting (16.5%, n=8).
Platelets increased the whole-clot permeability (Ks) by 3.5-fold and the whole-clot rigidity (G′) by 3-fold (Table⇓). These platelet-dependent modifications of the physical properties of the fibrin network were ascertained further by the reverse correlation (R) between Ks and G′ in PRCs (R=0.55, P=0.02) compared with FRCs (R=−0.64, P=0.0061; Figure 1⇑). It is likely that platelets within the PRC simultaneously strengthen and distort the fibrin network, resulting in more space within the pores delimited by fibrin fibers. This was confirmed by confocal microscopic analysis of the same clots. A typical FRC appeared as a homogeneous structure made up of straight rodlike elements corresponding to branching and crossing fibrin fibers organized in a 3D network (Figure 2A⇓). The round white elements correspond to branching fibers perpendicular to the plane of the scanning area. PRCs were rather inhomogeneous and were characterized by aggregates of platelets, for which the average surface area was 5562±521 μm2 (n=8, Figure 2B⇓). Fibrin fibers located within and at the edge of these aggregates were bent as if they were stressed and trapped. The double fluorescence labeling experiment showed that these aggregates corresponded to the colocalization of platelets and fibrin (Figure 2C⇓). Hence, superposition of fibrin-FITC and platelet-tricolor within the aggregates appeared as yellow areas, and FITC-fibrin alone appeared as green areas surrounding the platelet-fibrin aggregate (Figure 2C⇓).
Platelet-mediated fibrin retraction led to a 2.6-fold increase in the lysis time of PRCs compared with FRCs formed in the same conditions and with use of the static lysis assay (1 nmol/L rtPA, Figure 3a⇓). Raising the rtPA concentration up to 2 nmol/L lowered the average lysis time of PRCs by 32.8±20% (n=16, P<0.0001), but it still remained significantly higher than that of FRCs with 1 nmol/L of rtPA (Figure 3a⇓).
Effect of Adding Antiplatelet Agents Before Clotting
Unlike FRCs, mechanical properties of PRCs were greatly affected by the addition of abciximab or aspirin before the initiation of clotting (Table⇑). In all treated groups, a significant decrease in Ks and G′ was found. Unlike aspirin, abciximab displayed a dose-dependent efficacy in reducing G′ but not Ks. Unlike aspirin, abciximab (0.068 μmol/L) reversed the relation between Ks and G′, but in a nonsignificant manner (R=−0.28, P=0.28; Figure 1⇑). This relation was intermediate between the positive one of PRCs (R=0.55, P=0.02) and the negative one of FRCs (R=−0.64, P=0.0061). This indicates that despite a nearly complete inhibition of fibrinogen-platelet interactions, as shown by the measurement of GP IIb/IIIa receptor occupancy with the Accumetrics device, abciximab failed to restore a platelet-free–like fibrin structure. The significantly higher values of G′ and Ks of PRCs formed with 0.068 μmol/L of abciximab compared with FRCs further support this assumption. It seems likely that interactions between platelets and fibrin differ from interactions between platelets and fibrinogen.
These mechanical changes are correlated with morphological changes. Hence, adding abciximab or aspirin before PRC formation significantly decreased the average area of platelet-fibrin aggregates (S.ag), with a trend for a dose-dependent effect (Table⇑ and Figure 2D⇑ and 2E⇑). However, the only significant reduction between treated groups was found between 50 μg/mL of aspirin and 0.068 μmol/L of abciximab. These morphological changes resulted in a more homogeneous fibrin architecture with more fibrin exposed within the fibrin network, which is consistent with the concomitant reduction of both Ks and G′ (Table⇑).
Adding abciximab (0.034 μmol/L) and rtPA (1 nmol/L) before the initiation of clotting reduced the average lysis time of PRCs by 36.7±11% (P<0.001) compared with PRCs formed without abciximab. Addition of abciximab had the same effect as a 2-fold increase in the rtPA concentration (20.2±2.7 and 20.75±3.95 minutes, respectively; P=0.77; Figure 3a⇑). Although nonsignificant, there was a trend for a dose-related efficacy of the profibrinolytic effect of abciximab. The lysis rate of PRCs formed with 0.068 μmol/L abciximab still remained slower than the lysis rate of FRCs at a similar concentration of rtPA (P=0.019); this was consistent with the remaining difference in the physical properties between these 2 types of clots (Figure 3a⇑). Aspirin at 50 μg/mL increased the lysis rate of PRCs by 21±8% (n=8, P<0.01), but unlike abciximab, there was no trend for a dose-related effect (not shown).
Effects of Permeating Preestablished PRCs With Antiplatelet Agents
Permeation of preformed FRCs with aspirin or abciximab did not affect the physical properties of fibrin. Unlike buffer or aspirin, permeation of PRCs at a constant pressure gradient with 0.068 μmol/L abciximab significantly reduced Ks and G′, whereas S.ag was decreased in a nonsignificant manner (Table⇑ and Figure 2F⇑). Interestingly, these changes were found to be time-related. Early perfusion, starting within the 10 first minutes of clotting, was effective, whereas late perfusion, starting 40 minutes after the initiation of clotting, had no significant effect (Table⇑).
Permeation of preformed PRCs with rtPA (200 nmol/L) and aspirin (100 μg/mL) had an effect on the lysis time similar to that obtained by permeation with rtPA alone. Unlike aspirin, permeation of PRCs with abciximab (0.068 μmol/L) and rtPA (200 nmol/L) produced a 27% increase of the lysis speed (P<0.001, Figure 3b⇑). This accelerating effect of abciximab was not dose-related but time-related. As for mechanical properties, it was found with early permeation (starting within the first 10 minutes of initiation of clotting) but not with late permeation (starting after 40 minutes).
The present work was designed to evaluate the effect of platelets and platelet GP IIb/IIIa antagonists on fibrin physical properties and the fibrinolysis rate of native fully hydrated PRCs. It was hypothesized that inhibition of the interactions of platelet-membrane receptor GP IIb/IIIa with either fibrinogen or fibrin could modify the architecture of PRCs, leading to an acceleration of its dissolution. The most important finding of this in vitro study is that unlike aspirin, early permeation of preformed PRCs with abciximab can disaggregate these clots and facilitate fibrinolysis. These results provide new information on the mechanisms of abciximab to achieve full reperfusion of initially occluded coronary arteries.
Impact of Adding Platelets and Antiplatelet Agents Before Clotting on Physical Properties and Fibrinolysis Rate of PRCs
Our results provide for the first time direct correlations between mechanical and morphological properties of fully hydrated clots. Compared with FRCs, PRCs are stiffer, more porous (increase of G′ and Ks, respectively), and more heterogeneous, because of the presence of platelet-fibrin aggregate-like structures. These platelet-related changes of the fibrin physical properties account in part for the fibrinolysis resistance of PRCs. Adding antiplatelet agents before clotting prevented platelet-related mechanical (reduction of G′ and Ks) and morphological changes (reduction of S.ag) of the fibrin network, leading to a significant increase of the fibrinolysis rate. Unexpectedly, decreased permeability was found to be associated with increased lytic speed in the static lysis assay. This is because fibrin, which was trapped into the platelet-fibrin aggregates, has been reexposed in the surrounding fibrin network as a consequence of the inhibition of the interaction between fibrinogen and platelets by abciximab. Consequently, more fibrin is accessible to rtPA, which is loaded before the initiation of clotting, leading to a faster fibrinolysis, but the overall permeability of the clot is decreased.
The superiority of abciximab over aspirin in this setting was found to be directly related to a better inhibition of the fibrinogen-platelet interactions. Although aspirin provided a complete inhibition of the platelet cyclooxygenase activity, nearly all of the GP IIb/IIIa receptors remained functional after platelet activation with thrombin. The dose-related effect of abciximab that is reported in the present study further confirms previous in vivo pharmacodynamic studies showing a dose response of abciximab binding to platelet GP IIb/IIIa receptors.15 Interestingly, although only few GP IIb/IIIa receptors remained unoccupied with the highest dose of abciximab (0.068 μmol/L), a significant difference between the physical properties and the fibrinolysis rate of PRCs and FRCs was found. This finding is consistent with recent studies using recombinant fibrinogens showing that platelet interactions with fibrin may involve receptors other than GP IIb/IIIa.18 19 Furthermore, the present study corroborates previous data showing that at 0.034 to 0.068 μmol/L abciximab, platelet-mediated fibrin retraction is not completely abolished, although there is a complete inhibition of platelet aggregation at ≥0.034 μmol/L abciximab.20
Impact of Permeation of Antiplatelet Agents and/or rtPA on Physical Properties and Fibrinolysis Rate of Preformed PRCs
To consider rheological, transport, and enzymatic events, preformed PRCs were permeated at a constant gradient of pressure with antiplatelet agents and/or rtPA. As expected, PRCs were found more resistant to fibrinolysis than were FRCs. These findings are consistent with previous studies showing a decreased binding of tPA in PRCs as a consequence of platelet retraction.21 22 Unlike aspirin, permeation of PRCs with abciximab significantly decreased Ks and G′, suggesting that fibrin is removed from preexisting aggregates and reexposed into the surrounding fibrin network, as confirmed by confocal microscopy (decrease of S.ag). Interestingly, these abciximab-related changes of the fibrin network structure were found to be time-related and were observed only with early permeation (<10 minutes) and led to a faster fibrinolysis. Therefore, it is likely that the dissociating capacity of abciximab depends on the age of the PRCs and that the structure of the PRCs changes over time. This was ascertained by the continuous increase of G′ up to 40 minutes after the initiation of clotting and relates to the development of the platelet-contractile force that slowly retracts the fibrin network.23 Therefore, we assume that within the first 10 minutes of the initiation of clotting, the dissociating capacity of abciximab is strong enough to reverse the interactions between fibrin and activated platelets,24 whereas after 40 minutes, tight contacts between fibrin and platelets may prohibit the big molecule of abciximab (molecular weight of 47 000) to access the GP IIb/IIIa receptors. The fact that the interactions of fibrin with platelets may involve receptors other than GP IIb/IIIa, as shown above, is an important additional explanation for the limited efficacy of permeation of abciximab in disaggregating PRCs.
Relations of Our Experimental Design to Clinical Realities
Our data relate to clinical realities in different ways. Coronary thrombosis involves simultaneous deposition and lysis of red (fibrin-rich) and white (platelet-rich) clot components, causing intermittent vessel occlusion. Although this dynamic aspect of coronary thrombosis could not be directly assessed in our experimental design, our data clearly emphasize that abciximab could affect simultaneously preformed PRCs (disaggregating potential) and clot extension by the prevention of platelet-related fibrin remodeling. Moreover, it is likely that the potential of abciximab to reduce the generation of thrombin may accentuate the limitation of the clot extension in vivo, although this has not been directly demonstrated in our work.25 Hence, fibrin expression from platelets plays a minor role in our experimental system because fibrin formation is mainly related to the activation of fibrinogen molecules by the exogenous thrombin addition. Thrombin that is generated as a consequence of platelet activation has a different kinetic profile and usually appears after the activation of nearly all the fibrinogen molecules.
The time-related efficacy of abciximab in disaggregating preformed PRCs is consistent with the temporal relationship detected between the age of the thrombus and the efficacy of abciximab in reperfusing the occluded coronary artery in vivo.26 These in vitro data further support recent results of randomized trials showing that abciximab with reduced-dose fibrinolytic agents enhances fibrin and platelet lysis, resulting in rapid and complete reperfusion (within 60 minutes) in a high proportion of patients with myocardial infarction (nearly 70%).4 5 Moreover, our findings are consistent with those of the same trials showing that adding low-dose reteplase to abciximab provides slightly superior rates of culprit artery patency compared with those with standard-dose reteplase alone (see Figure 3⇑).
In terms of clot mechanics, final platelet count is obviously a critical parameter and therefore a limitation of our experimental design. We found a correlation between clot strength and platelet count with a dose-related enhancement of G′, as described earlier: G′=420+p0.76 (where p is platelet concentration).27 In parallel, the whole-clot permeability, Ks, was found to increase linearly with platelet concentration. However, accurate measurements of Ks were impaired at >100 000 platelets per microliter because retraction led to the formation of leaks. For the same reasons, morphological analyses were not possible. Fibrinolysis resistance using the static lysis assay was found to be positively correlated with the platelet concentration, and the effect of abciximab on physical properties and the fibrinolysis rate was of a similar magnitude over the whole range of platelet concentration.
The question arises as to whether our experimental conditions are relevant to an in vivo scenario, especially regarding the retraction phenomena. Red blood cells are an important component of the thrombus, primarily because their incompressibility limits clot retraction to 50%.13 In our system, red blood cells were discarded because the system did not allow optical microscopy to be performed; however, the high surface area of contact of the clots within the microchamber considerably limited the retraction phenomena. Therefore, we assume that a final platelet count of 75 000 per microliter may have a similar impact on the fibrin network as a platelet concentration of 150 000 per microliter in the presence of red blood cells.
Although the present study emphasizes the role of PRC architecture as an important determinant of the fibrinolysis rate, other factors are of major importance. Plasminogen activator inhibitor type 1, which has been shown to be a major determinant of the failure of thrombolytic therapy in vitro and in vivo, may have an important role in our experimental setting.28 29 30 It has been detected only in platelet-fibrin aggregates and with a higher intensity in PRCs formed with buffer compared with abciximab. Further quantitative analyses are needed to evaluate the impact of plasminogen activator inhibitor type 1 in our setting. A structural and dynamic approach to the early stage of PRC formation and fibrinolysis is also an important issue and is also needed to evaluate our findings and to extend previous work on the concept of a heterogeneous architecture of occlusive PRCs with aggregates that resist fibrinolysis.31 32 33
In conclusion, inhibition of platelet-mediated fibrin remodeling and removal of fibrin from preexisting aggregates contribute to the disaggregating potential of abciximab and are potential mechanisms for its additional benefit over aspirin in the setting of thrombolysis in myocardial infarction.
This study was supported by a grant from the Association de Cardiologie Ile-de-France, Paris, France. We wish to thank J.W. Weisel, PhD, from the Department of Cell and Developmental Biology, University of Pennsylvania (Philadelpha, Pa), for the fruitful discussion and his expert review of the manuscript.
Gold HK, Garabedian HD, Dinsmore RE, Guerrero LJ, Cigarroa JE, Palacios IF, Leinbach CL. Restoration of coronary flow in myocardial infarction by intravenous chimeric 7E3 antibody without plasminogen activators. Circulation. 1997;95:1755–1759.
Montalescot G, on behalf of the ADMIRAL investigators. Abciximab associated with primary angioplasty and stenting in acute myocardial infarction: the ADMIRAL Study, 30-day final results. In: 72nd Scientific Sessions of the American Heart Association; November 7–10, 1999; Atlanta, Ga. Abstract 446.
Brener SJ, Barr LA, Burchenal JE, Katz S, George BS, Jones AA, Cohen ED, Gainey PC, White HJ, Cheek HB, et al. Randomized, placebo-controlled trial of platelet glycoprotein IIb/IIIa blockade with primary angioplasty for acute myocardial infarction: ReoPro and Primary PTCA Organization and Randomized Trial (RAPPORT) Investigators. Circulation. 1998;98:734–741.
Antman EM, Giugliano RP, Gibson CM, McCabe CH, Coussement P, Kleiman NS, Vahanian A, Adgey AAJ, Menown I, Rupprecht HR, et al. Abciximab facilitates the rate and extent of thrombolysis: results of the Thrombolysis in Myocardial Infarction (TIMI) 14 Trial. Circulation. 1999;99:2720–2732.
The Speed Investigators. Trial of abciximab with and without low-dose reteplase for acute myocardial infarction. Circulation. 2000;101:2788–2794.
Beucher S, Meyer F. The morphological approach to segmentation: the watershed transformation. In: Dougherty ER, ed. Mathematical Morphology in Image Processing. Rochester, NY: Marcel Dekker; 1993:433–482.
Russ JC. The Image Processing Handbook. 2nd ed. Boca Raton, Fla/Ann Arbor, Mich/London, UK/Tokyo, Japan: CRC Press; 1995.
Shen LL, McDonagh RP, McDonagh J, Hermans J. Early events in the plasmin digestion of fibrinogen and fibrin: effects of plasmin on fibrin polymerization. J Biol Chem. 1977;252:6184–6189.
Cipollone F, Patrignani P, Greco A, Panara MR, Padovano R, Cuccurullo F, Patrono C, Rebuzzi AG, Liuzzo G, Quaranta G, et al. Differential suppression of thromboxane biosynthesis by indobufen and aspirin in patients with unstable angina. Circulation. 1997;96:1109–1116.
Gold HK, Gimple LW, Yasuda T, Leinbach RC, Werner W, Holt R, Jordan R, Berger H, Collen D, Coller BS. Pharmacodynamic study of (Fab′)2 fragments of murine monoclonal antibody 7E3 directed against human platelet glycoprotein IIb/IIIa in patients with unstable angina. J Clin Invest. 1990;86:651–659.
Coller BS, Lang DL, Scudder LE. Rapid and simple platelet function assay to assess glycoprotein IIb/IIIa receptor blockade. Circulation. 1997;95:860–867.
Rooney MM, Parise LV, Lord ST. Dissecting clot retraction and platelet aggregation. J Biol Chem. 1996;271:8553–8555.
Rooney MM, Farrell DH, van Hemel BM, de Groot PG, Lord ST. The contribution of the three hypothesized integrin-binding sites in fibrinogen to platelet-mediated clot retraction. Blood. 1998;92:2374–2381.
Carr ME, Carr SL, Hantgan RR, Braaten J. Glycoprotein IIb/IIIa blockade inhibits platelet-mediated force development and reduces gel elastic modulus. Thromb Haemost. 1995;73:99–106.
Jang IK, Gold HK, Ziskind AA, Fallon JT, Holt RE, Leinbach RC, May JW, Collen D. Differential sensitivity of erythrocyte-rich and platelet-rich arterial thrombi to lysis with recombinant tissue-type plasminogen activator: a possible explanation for resistance to coronary thrombolysis. Circulation. 1989;79:920–928.
Kunitada S, Fitzgerald GA, Fitzgerald DJ. Inhibition of clot lysis and decreased binding of tissue-type plasminogen activator as a consequence of clot retraction. Blood. 1992;81:1420–1427.
Carr ME, Zeckert SL. Measurement of platelet mediated force development during plasma clot formation. Am J Sci. 1991;302:13–19.
Scarborough RM, Kleiman S, Phillips DR. Platelet glycoprotein IIb/IIIa antagonists: what are the relevant issues concerning their pharmacology and clinical use? Circulation. 1999;100:437–444.
Reverter JC, Béguin S, Kessels H, Kumar R, Hemker HC, Coller BS. Inhibition of platelet-mediated, tissue-factor-induced thrombin generation by the mouse/human chimeric 7E3 antibody: potential implications for the effect of c7E3 Fab treatment on acute thrombosis and “clinical restenosis.” J Clin Invest. 1996;98:1284–1291.
Mahdi NA, Leinsbach R, Dinsmore RE, Newell J, Gold HK. Predictors of coronary reflow with intravenous chimeric 7E3 antibody in acute myocardial infarction. J Am Coll Cardiol. 1998;31(suppl A):230A. Abstract.
Braaten JV, Handt S, Jerome W, Kirkpatrick J, Lewis JC, Hantgan RR. Regulation of fibrinolysis by platelet released plasminogen activator inhibitor 1: light scattering and ultrastructural examination of lysis of a model platelet fibrin thrombus. Blood. 1993;81:1290–1298.
Braaten JV, Gray J, Hantgan RR. Uncoupling fibrin from integrin receptors hastens fibrinolysis at the platelet-fibrin interface. Blood. 1994;83:982–993.
Zhu Y, Carmeliet P, Fay WP. Plasminogen activator inhibitor-1 is a major determinant of arterial thrombolysis resistance. Circulation. 1999;99:3050–3055.
Bugelski PJ, Kopia GA, Kopiaciewicz L, Cadogan AS, Morgan DG. Ultrastructural analysis of thrombolysis by streptokinase and tissue-type plasminogen activator of experimental coronary artery thrombosis. Fibrinolysis. 1989;3:137–144.
Blinc A, Kennedy SD, Bryant RG, Marder VJ, Francis CW. Flow through clots determines the rate and pattern of fibrinolysis. Thromb Haemost. 1994;71:330–337.