Effects of Reteplase and Alteplase on Platelet Aggregation and Major Receptor Expression During the First 24 Hours of Acute Myocardial Infarction Treatment

View metadata, citation and similar papers at core.ac.uk brought to you by CORE

provided by Elsevier - Publisher Connector

1466 JACC Vol. 31, No. 7 June 1998:1466–73

PAUL A. GURBEL, MD, FACC, VICTOR L. SEREBRUANY, MD, PHD, ANDREW R. SHUSTOV, MD, RAYMOND D. BAHR, MD, FACC, CYNTHIA CARPO, MD, E. MAGNUS OHMAN, MD, FACC,* ERIC J. TOPOL, MD, FACC,† FOR THE GUSTO-III INVESTIGATORS‡ Baltimore, Maryland; Durham, North Carolina; and Cleveland, Ohio expression at 24 h. Trends toward (0.002 ؍ Objectives. We sought to compare platelet characteristics after molecule-1 (p reteplase and alteplase therapy in the setting of the Global Use of decreased receptor expression early (3 to 6 h), followed by a Strategies to Open Occluded Coronary Arteries (GUSTO)-III trial. progressive increase at 12 h and especially at 24 h occurred after Background. Platelet function may be impaired during throm- both agents. bolysis in patients with an acute myocardial infarction. The effects Conclusions. In this prospective clinical ex vivo platelet study, of reteplase and alteplase on platelet aggregation and major similar patterns of platelet aggregation and surface receptor surface antigen expression during the first 24 h of infarction expression occurred during the first 24 h of coronary thrombolysis therapy are unknown. with reteplase and alteplase. However, after reteplase, indicators Methods. Platelet aggregation and receptor expression by flow of platelet activity were higher at 24 h after thrombolysis than cytometry were determined in 23 patients before thrombolysis and after alteplase. These data suggest that GP IIb/IIIa inhibitors or thereafter at 3, 6, 12 and 24 h. other antiplatelet strategies may be particularly advantageous Results. Aggregation was higher after reteplase at 24 h when when used 12 to 24 h after thrombolysis, especially after reteplase therapy. It is at this time point during the first day of coronary ؍ induced by 5 ␮mol/liter adenosine diphosphate (ADP) (p and thrombolysis that GP IIb/IIIa is markedly expressed and platelets (0.003 ؍ collagen (p ,(0.02 ؍ ␮mol/liter ADP (p 10 ,(0.007 .than after alteplase. Reteplase therapy are most active (0.009 ؍ thrombin (p (very late (J Am Coll Cardiol 1998;31:1466–73 ,(0.04 ؍ exhibited greater glycoprotein (GP) IIb/IIIa (p and platelet/endothelial cell adhesion ©1998 by the American College of Cardiology (0.04 ؍ antigen-2 (p

Thrombolytic therapy is the standard treatment for patients (9–11), combined with thrombolytic agents, have been shown with acute myocardial infarction (AMI). Whereas streptoki- in animal studies and clinical trials to improve reperfusion. nase and alteplase are established agents, the clinical efficacy The attempt to determine the platelet-related properties of of reteplase is evolving. Platelets play an important role in the thrombolytic agents is not new. There is substantial evidence natural history of AMI. Intravascular platelet activation may from in vitro (12–14) and animal studies (15,16) that impaired limit reperfusion or cause reocclusion of the recanalized platelet function occurs after thrombolytic therapy. Although infarct-related arteries, thus resulting in an overall decreased there is agreement that concentrations of substances released effectiveness of thrombolytic therapy (1,2). Glycoprotein (GP) by platelets are elevated after thrombolysis (17–20), there is IIb/IIIa receptor antagonists (3–8) and monoclonal antibodies no consensus about the dynamics of ex vivo platelet status in this setting (21–24). Such discrepancies could relate partly to the particular thrombolytic agent, limitations of the techniques to assess platelet function, the lack of a standard From the Union Memorial Hospital and St. Agnes Hospital, Baltimore, method to sample blood and no specified time course to Maryland; *Duke Clinical Research Institute, Durham, North Carolina; and handle samples. Indeed, the majority of human myocardial †Department of Cardiology, Cleveland Clinic, Cleveland, Ohio. ‡A complete list of the GUSTO-III Investigators appears in reference 27. This study was infarction (MI) studies lack a careful, precise description of supported in part by Boehringer Mannheim GmbH, Mannheim, Germany and by the time course of platelet monitoring. The random mea- Medtronic, Inc., San Diego, California and was presented in part at the 70th surement of platelet function, outside strict time intervals Scientific Sessions of the American Heart Association, Orlando, Florida, No- vember 1997. during MI, has revealed very limited information about Manuscript received September 22, 1997; revised manuscript received March arterial patency, thrombolytic success and possible platelet 2, 1998, accepted March 5, 1998. involvement in reocclusion, reinfarction and bleeding. Many Address for correspondence: Dr. Paul A. Gurbel, Center for Thrombosis Research, Sinai Hospital of Baltimore, 2401 West Belvedere Avenue, Research clinical studies have examined platelets at unspecified times Building, R202, Baltimore, Maryland 21215. E-mail: [email protected]. or even days after thrombolysis (17,20,22–25), when the

©1998 by the American College of Cardiology 0735-1097/98/$19.00 Published by Elsevier Science Inc. PII S0735-1097(98)00172-7 JACC Vol. 31, No. 7 GURBEL ET AL. 1467 June 1998:1466–73 PLATELETS DURING THROMBOLYSIS IN AMI

infusion of heparin for the first 24 h after thrombolysis, as Abbreviations and Acronyms recommended in the GUSTO-III protocol. Eighteen patients ADP ϭ adenosine diphosphate had successful reperfusion and remained free of recurrent AMI ϭ acute myocardial infarction ischemia in the first 24 h of their hospital stay. Three patients ϭ FITC fluorescein isothiocyanate (two with reteplase, one with alteplase) had persistent chest GP ϭ glycoprotein GUSTO ϭ Global Use of Strategies to Open Occluded Coronary pain and ST elevation and underwent immediate angiography, Arteries which revealed the absence of reperfusion. Two patients (one MI ϭ myocardial infarction with reteplase, one with alteplase) developed recurrent isch- PAI-1 ϭ plasminogen activator inhibitor-1 emia in the first 24 h and also underwent emergency angiog- ϭ PECAM-1 platelet/endothelial cell adhesion molecule-1 raphy. TBS ϭ Tris-buffered saline VLA-2 ϭ very late antigen-2 Time course and exclusion of blood samples. The sched- ules for blood drawing, sample preparation and processing were critical issues of the study design and were monitored by an independent observer (A.R.S.). The actual timing of blood effects of adjunctive therapy cannot be excluded (26). collection for the baseline sample was 9.5 Ϯ 1.4 min (mean Ϯ Moreover, our knowledge of the status of platelets during SD) before the start of thrombolytic therapy, 174.6 Ϯ 21.8 min the first 24 h after thrombolysis is very limited, especially for the 3-h sample, 371.1 Ϯ 24.2 min for the 6-h sample, from controlled clinical trials. 709.4 Ϯ 17.8 min for the 12-h sample and 1,402.9 Ϯ 18.8 min The purpose of the present study was to define the imme- for the 24-h sample. Samples were processed within 1 h after diate and early platelet-related effects of reteplase and alte- blood drawing. Four patients (three in the reteplase group, one plase in AMI. We assessed platelet aggregation in response to in the alteplase group) did not complete the protocol at all multiple agonists and determined major surface receptor ex- time points. The reasons for early termination were patient pression with flow cytometry at specified times after thrombol- transfer for emergency angioplasty (n ϭ 3) and inability to ysis in patients enrolled in the randomized Global Use of obtain a blood sample (n ϭ 1). Twenty-three baseline samples, Strategies to Open Occluded Coronary Arteries (GUSTO-III) 22 samples collected at 3 h, 20 samples collected at 6 h, 20 trial. samples collected at 12 h and 19 samples collected at 24 h were included in the analysis. Platelet aggregation. Blood samples for platelet aggrega- Methods tion and flow cytometric studies were taken at specified The study was approved by the Institutional Review Board intervals: in the emergency department immediately before of St. Agnes Hospital and Union Memorial Hospital (Balti- administration of the thrombolytic therapy and in the coronary more, Maryland). Written informed consent was obtained care unit at 3, 6, 12 and 24 h after thrombolytic therapy began. from all study participants. Citrate and whole blood were immediately mixed in a 1:9 Patients. Twenty-three consecutive patients admitted to ratio and centrifuged at 1,200 ϫ g for 2.5 min. The resulting the emergency departments of St. Agnes Hospital or Union platelet-rich plasma was kept at room temperature for use Memorial Hospital between July and December 1996 with a within 1 h. Platelet counts were determined for each plasma diagnosis of AMI were included. All patients were enrolled in sample with a Coulter counter ZM. Platelet numbers were the GUSTO-III trial of reteplase versus accelerated alteplase adjusted to 3.50 ϫ 108/ml with homologous platelet-poor treatment for AMI. The inclusion criteria have been reported plasma. Platelet aggregation was induced by 5 and 10 ␮mol/ elsewhere (27). In summary, patients of any age who presented liter adenosine diphosphate (ADP), 1 ␮g/ml of collagen, 1 within6hofsymptom onset with Ͼ30 min of continuous mg/ml of thrombin and 1.25 mg/ml of ristocetin. All agonists symptoms of AMI and had Ն1-mm ST segment elevation in were obtained from Chronolog Corporation. Aggregation two or more limb leads, Ն2-mm ST segment elevation in two studies were performed with a Chronolog Whole Blood Lumi- or more contiguous precordial leads or bundle branch block on Aggregometer (Model 560-Ca). Aggregation was expressed as the 12-lead electrocardiogram were eligible for enrollment. the maximal percent change in light transmittance from base- Patients were excluded for bleeding diathesis; previous stroke, line, using platelet-poor plasma as a reference at the end of the major operation or significant trauma in the past 6 weeks; or recording time. Aggregation curves were recorded for 4 min hypertension Ͼ200/110 mm Hg. and analyzed according to international standards (28). Treatments and clinical outcomes. Patients randomized to Flow cytometry. Flow cytometric procedures have been receive reteplase received two 10-MU boluses given 30 min described elsewhere (29,30) and were performed according to apart. Patients assigned to alteplase received an accelerated the manufacturer’s recommendations for sample preparation regimen: a 15-mg bolus, then 0.75 mg/kg body weight over and processing. Briefly, venous blood (8 ml) was collected in a 30 min, then 0.50 mg/kg over 1 h. All patients in the current plastic tube containing 2 ml of acid-citrate-dextrose (7.3 g of ϫ substudy had received 325 mg of aspirin and at least 5,000 U of citric acid, 22.0 g of sodium citrate 2H2O and 24.5 g of intravenous heparin at baseline sampling. After administration glucose in 1 liter of distilled water) and mixed well. The of thrombolytic therapy, all patients received a continuous mixture was centrifuged at 1,000 rpm for 10 min at room 1468 GURBEL ET AL. JACC Vol. 31, No. 7 PLATELETS DURING THROMBOLYSIS IN AMI June 1998:1466–73 temperature. The upper two-thirds of the platelet-rich plasma Table 1. Clinical Characteristics of 23 Patients With an Acute was then collected and adjusted to a pH of 6.5 by adding Myocardial Infarction acid-citrate-dextrose. It was then centrifuged at 1,600 ϫ g for Reteplase Alteplase 10 min. The supernatant was removed, and the platelet pellet (n ϭ 13) (n ϭ 10) was gently resuspended in 4 ml of washing buffer (10 mmol/ Male/female 10/3 8/2 liter Tris/HCl, 0.15 mol/liter NaCl, 20 mmol/liter EDTA, pH Age (yr) 59.4 Ϯ 9.5 62.7 Ϯ 8.8 7.4). Platelets were washed in the washing buffer and then in Range 38–79 36–86 Tris-buffered saline (TBS) (10 mmol/liter Tris, 0.15 mol/liter Current smoker 6 5 NaCl, pH 7.4). All cells were then divided into 10 plastic Alcohol use 3 3 Hypertension 8 6 capped tubes. Nine portions of washed platelets were incu- ␮ Diabetes 3 2 bated with 5 l of fluorescein isothiocyanate (FITC)– Prior MI 3 1 conjugated antibodies in darkness at 4°C for 30 min, and one Hypercholesterolemia 5 4 part remained unstained and served as a negative control. Baseline medications Surface antigen expression was measured with monoclonal CCBs 3 3 murine antihuman antibodies to CD9 (p24), CD41 (GP IIb/ BBs 2 2 IIIa, ␣ ␤ ), CD42b (GP Ib), CD61 (GP IIIa) (DAKO Cor- ACEIs 1 0 IIb 3 Nitrates 2 1 poration), CD49b (very late antigen [VLA]-2, or ␣ ␤ ), CD62p 2 1 Diuretic drugs 1 1 (P-selectin), CD31 (platelet/endothelial cell adhesion molecule Aspirin 5 2 ␣ ␤ [PECAM]-1), CD 41b (gp IIb), and CD51/CD61 ( v 3,or Ant/inf MI 8/5 6/4 vitronectin receptor) (PharMingen). The antibody to the GP Laboratory data IIb/IIIa receptor has been demonstrated to react with the WBCs (ϫ103/ml) 8.2 Ϯ 1.6 7.9 Ϯ 1.9 platelet population at rest and exhibits a dose-dependent RBCs (ϫ106/ml) 3.76 Ϯ 0.5 3.59 Ϯ 0.5 ϫ 3 Ϯ Ϯ increase of receptor binding after ADP or thrombin stimula- Platelets ( 10 /ml) 225.1 11.7 238.4 10.5 Hemoglobin (g/dl) 15.2 Ϯ 0.6 15.4 Ϯ 0.5 tion, or both. After incubation, the cells were washed in TBS Hematocrit (%) 40.9 Ϯ 3.8 43.7 Ϯ 3.9 and resuspended in 0.25 ml of 1% paraformaldehyde. Samples Creatinine (mg/dl) 1.0 Ϯ 0.3 0.9 Ϯ 0.4 were stored in the refrigerator at 4°C and analyzed with a Time from symptom onset 214.5 Ϯ 19.8 208.22 Ϯ 21.0 Becton Dickinson FACScan flow cytometer with laser output to thrombolysis (min) of 15 mW, excitation at 488 nm and emission detection at Data presented are mean value Ϯ SD, range or number of patients. ACEIs ϭ 530 Ϯ 30 nm. The instrument was calibrated daily with angiotensin converting enzyme inhibitors; Ant ϭ anterior; BBs ϭ beta-blockers; fluorescence beads (CaliBRITE, Becton Dickinson) and mea- CCBs ϭ calcium blockers; inf ϭ inferior; MI ϭ myocardial infarction; RBCs ϭ sured FITC-conjugated fluorescence intensity. All variables red blood cells; WBCs ϭ white blood cells. were obtained using four-decade logarithmic amplification. The data were collected, stored in list mode and then analyzed Platelet aggregation. Baseline platelet function did not using CELLQuest (version 1.2.2) software. differ between groups. In patients treated with alteplase, Statistical analysis. A post hoc comparison using the t test platelet aggregability did not differ significantly between the with the Bonferroni correction was performed to identify baseline sample and that at any other time point in response to specific differences in platelet aggregation and receptor expres- all agonists. However, reteplase-treated patients revealed sig- sion between time points within each treatment group. nificant changes in platelet aggregation induced by all agonists Between-treatment comparisons were made at each time point except ristocetin (Fig. 1). Platelets aggregated significantly using t tests (platelet aggregation). A Mann-Whitney U test more in response to 5 ␮mol/liter ADP in the reteplase group at was used to analyze nonparametric data (flow cytometry). 12h(pϭ 0.04) and 24 h (p ϭ 0.007) than in the alteplase Normally distributed data are expressed as mean value Ϯ SD. group. Compared with baseline values, platelet aggregability Data distributed otherwise are summarized as medians was significantly reduced at3h(pϭ 0.02) and6h(pϭ 0.02) (range), and p Ͻ 0.05 was considered significant. Differences after reteplase therapy. A similar pattern occurred in response between individual flow cytometric histograms were assessed to 10 ␮mol/liter ADP. Platelets were significantly more active using the Smirnov-Kolmogorov test incorporated in the in the reteplase group after 24 h (p ϭ 0.02) than in the CELLQuest software. alteplase group, and platelet aggregation was significantly reduced3h(pϭ 0.03) and6h(pϭ 0.04) after reteplase treatment compared with baseline values. In response to the Results collagen agonist, platelet aggregation was again significantly Baseline clinical and demographic characteristics did not increased in the reteplase group at 24 h (p ϭ 0.003) compared differ substantially between the reteplase and alteplase groups with that in the alteplase group. Significant reduction of (Table 1). Three more patients received aspirin daily in the platelet aggregability occurred6h(pϭ 0.009) after reteplase reteplase group. Groups were also similar in location of therapy compared with baseline values. As with collagen infarction and other baseline clinical and laboratory character- stimulation, the only difference observed between groups after istics before thrombolysis. thrombin inducement was at 24 h, when platelets aggregated JACC Vol. 31, No. 7 GURBEL ET AL. 1469 June 1998:1466–73 PLATELETS DURING THROMBOLYSIS IN AMI

significantly more in the reteplase group (p ϭ 0.01). With Figure 1. Platelet aggregation in response to administration of 5 ␮ ␮ ristocetin, platelet aggregation was consistently high, ranging mol/liter ADP (A), 10 mol/liter ADP (B), collagen (C) and throm- (D) (circles) between 92.6% and 94.1% for reteplase and between 92.7% bin in patients treated with reteplase versus alteplase (squares) for AMI. *p Ͻ 0.05 between groups. and 93.1% for alteplase during the first 24 h after thrombolysis, with no significant differences between and within groups. Flow cytometry. Baseline platelet receptor expression did 24 h for reteplase (p ϭ 0.002) and alteplase (p ϭ 0.035) not differ significantly between the reteplase and alteplase treatment, but significantly more so in the reteplase group (p ϭ groups, but substantial changes in receptor expression resulted 0.037), which exhibited a nearly threefold increase compared after thrombolysis in both groups (Table 2, Fig. 2). Expression with baseline values and was twofold higher than that in the of p24 did not differ significantly between groups at any time alteplase group. Representative overlapped histograms of point. Significant decreases in receptor expression compared platelets labeled with fluorescent antibodies to GP IIb/IIIa with baseline values were seen at3h(pϭ 0.004) and6h(pϭ from patients treated with reteplase or alteplase are presented 0.005), followed by an increase (p ϭ 0.04) at 24 h, after in Figure 3. ␣ ␤ reteplase therapy. Within the alteplase group, no early recep- The only significant difference in platelet VLA-2 ( 2 1) tor inhibition was observed, but at 24 h receptor expression was expression between and within groups was an almost threefold higher (p ϭ 0.01) than that at baseline. Similar receptor increase in fluorescent intensity among reteplase-treated pa- expression patterns were seen with GP Ib in both groups, with tients at 24 h (p ϭ 0.04) compared with alteplase-treated significant inhibition 12 h after either reteplase (p ϭ 0.02) or patients. alteplase (p ϭ 0.007) therapy, but there were no significant Very similar profiles in P-selectin expression were observed differences between groups. between groups: Early significant decreases (p ϭ 0.001 for the Although a slight early decrease, followed by an increase, in reteplase group; p ϭ 0.009 for the alteplase group) were GP IIb occurred in both thrombolytic groups, there were no followed by significant increases in its expression (p ϭ 0.009 for significant differences between or within the groups. A signif- the reteplase group; p ϭ 0.02 for the alteplase group). icant increase in GP IIIa expression was observed at 12 h (p ϭ For PECAM-1, a significant increase (p ϭ 0.002) in fluo- 0.01) and 24 h (p ϭ 0.0003) after reteplase, but not alteplase, rescent intensity occurred in the reteplase group after 24 h therapy. Again, these differences were not significant between compared with the alteplase group. A significant decrease in its groups. expression (p ϭ 0.01) was seen 3 h after reteplase treatment ␣ ␤ Compared with baseline values, GP IIb/IIIa ( IIb 3) ex- compared with baseline values. ϭ ␣ ␤ pression was decreased significantly (p 0.03) at 3 h after Dynamic changes in platelet vitronectin receptor ( V 3) reteplase therapy. Its expression was markedly increased at expression were similar between groups; a significant increase 1470 GURBEL ET AL. JACC Vol. 31, No. 7 PLATELETS DURING THROMBOLYSIS IN AMI June 1998:1466–73

Table 2. Platelet Surface Antigen Expression in Patients With Acute Myocardial Infarction During First 24 Hours After Thrombolytic Therapy Group Baseline 3 h 6 h 12 h 24 h CD9 Reteplase 53.31 Ϯ 18.41 41.12 Ϯ 13.68* 43.69 Ϯ 13.46* 48.62 Ϯ 19.73 83.39 Ϯ 58.33* Alteplase 57.58 Ϯ 16.28 53.40 Ϯ 11.66 58.15 Ϯ 15.45 61.14 Ϯ 15.32 75.20 Ϯ 30.28* GP IIb Reteplase 28.76 Ϯ 3.08 26.92 Ϯ 5.89 27.62 Ϯ 5.75 29.77 Ϯ 5.64 29.39 Ϯ 5.46 Alteplase 27.96 Ϯ 1.55 27.92 Ϯ 3.60 28.70 Ϯ 5.03 29.71 Ϯ 4.47 31.25 Ϯ 5.47 GP IIIa Reteplase 318.29 Ϯ 48.28 309.54 Ϯ 65.01 339.92 Ϯ 65.36 349.77 Ϯ 64.52* 385.23 Ϯ 62.68* Alteplase 338.25 Ϯ 56.52 323.40 Ϯ 67.76 329.90 Ϯ 82.97 347.36 Ϯ 41.04 369.50 Ϯ 45.11 P-selectin Reteplase 31.78 Ϯ 4.64 29.90 Ϯ 4.68* 30.63 Ϯ 3.48 32.54 Ϯ 4.26 33.92 Ϯ 4.05* Alteplase 31.23 Ϯ 5.59 29.60 Ϯ 5.17* 30.00 Ϯ 5.21 30.50 Ϯ 5.84 32.20 Ϯ 5.03* Vitronectin receptor Reteplase 59.41 Ϯ 22.18 53.62 Ϯ 30.49 56.04 Ϯ 33.68 73.69 Ϯ 44.05* 92.23 Ϯ 58.54* Alteplase 61.12 Ϯ 30.64 53.70 Ϯ 26.88 60.80 Ϯ 32.11 75.30 Ϯ 25.55* 68.11 Ϯ 27.69 *p Ͻ 0.05 versus corresponding baseline measurement. Data presented are mean value Ϯ SD, expressed as log ﬂuorescence intensity.

occurred at 12 h after thrombolysis with reteplase (p ϭ 0.04) or result in reduced thrombolysis. Specific binding sites for plas- alteplase (p ϭ 0.04). At 24 h after thrombolysis, vitronectin minogen (36) and tissue-type plasminogen activator (37) have receptor expression was even greater in the reteplase group been described on the platelet surface. The localization of (p ϭ 0.008), whereas its expression in the alteplase group these mediators of fibrinolysis to platelets suggests that they trended toward baseline levels. may play an important role in the modulation of clot lysis. The effect of thrombolytic therapy on platelets has been extensively debated, with evidence of both platelet activation Discussion and inhibition. Most studies have used in vitro models and We determined the ex vivo platelet-related properties of indirect measures of activity, such as the determination of two thrombolytic agents in patients with an AMI with the levels of various mediators and other constituents released simultaneous use of flow cytometry and conventional aggrega- from platelets. The recent use of flow cytometry has enhanced tion techniques. We found that reteplase and alteplase, as used our knowledge of events that occur on the platelet surface in the GUSTO-III trial, have similar effects on platelet aggre- during activation, allowing the precise measurement of recep- gation in response to four agonists. The groups differed only tors expressed at specified times after thrombolysis. However, late after administration (Ͼ12 h), in that the reteplase group there are few prospective studies of platelet flow cytometry in showed enhanced aggregability in response to ADP, collagen patients with an AMI. and thrombin. In addition, the reteplase group showed mild Platelet aggregation. In the alteplase group, no significant early dysfunction (at 3 to 6 h) that was followed by late changes in aggregation occurred after treatment. These data enhanced aggregation. We also observed significant changes in are similar to the results of Bertolino et al. (21), who observed the expression of multiple surface antigens after treatment no difference in aggregation in response to 5 ␮mol/liter ADP with both agents, particularly GP IIb/IIIa, VLA-2 and or collagen at 24 h after conventional dosing of alteplase. PECAM-1 at 24 h. However, the significant reduction in aggregation (ϳ40%) at 1 Platelets and thrombolytic therapy. Failure to reperfuse, to 3 and at 6 h after drug administration contrasts with our acute reocclusion and severe bleeding complications are major findings. Ristocetin-induced aggregation was not affected by limitations of systemically given thrombolytic therapy (31,32). alteplase in either study. To our knowledge, there are no Platelet thrombus formation has been implicated in all these published data on the time course of ex vivo platelet aggrega- events. Thus, the effect of thrombolytic therapy on platelet tion in patients with an AMI treated with reteplase. characteristics are important. Vasoconstriction affected by Receptor expression. In the current study, we saw signifi- platelet release of thromboxane and other mediators, dimin- cant changes in the expression of multiple surface antigens ished heparin effects induced by platelet factor 4 and enhanced after treatment with both agents. Early after thrombolysis, we thrombin generation on the platelet surface all adversely affect observed only minor (Ͻ20%), antigen-specific reductions in reperfusion (33). Platelets also interact with both physiologic receptor expression, which could reflect excessive plasmin- and drug-induced fibrinolysis. For example, platelet granules induced cleavage or occupancy of the receptor by ligands (38). contain plasminogen activator inhibitor-1 (PAI-1) and alpha2- The earliest decreases in antigen expression occurred with antiplasmin (34,35). Release of these proteins can theoretically platelet antigen 24, GP IIb/IIIa, P-selectin and PECAM-1. JACC Vol. 31, No. 7 GURBEL ET AL. 1471 June 1998:1466–73 PLATELETS DURING THROMBOLYSIS IN AMI

Later in thrombolysis (Ͼ12 h), we observed a marked increase Figure 2. Platelet receptor expression in patients treated with rete- in specific receptor expression, which was pronounced for plase (circles) versus alteplase (squares) for AMI: GP Ib (A), GP (B), (C) (D). Ͻ VLA-2, PECAM-1 and GP IIb/IIIa. Bihor et al. (22) studied IIb/IIIa VLA-2 and PECAM-1 *p 0.05 between groups. three patients treated with alteplase by platelet flow cytometry using monoclonal antibodies to P-selectin and fibrinogen bound to GP IIb/IIIa (22). They described GP IIb/IIIa activa- platelet aggregability and receptor expression after reteplase tion within the first 72 h after treatment, which is in agreement could be relevant to early reperfusion. However, the GUSTO- with our findings. III mortality data did not differ between agents (27). This A decrease in platelet GP Ib expression, observed after 12 h finding may be explained in part by increased platelet activa- of reteplase or alteplase therapy, supports previous in vitro tion observed in the reteplase group later at 24 h after reports that thrombolysis attenuates its expression (39,40) and thrombolysis. An intriguing postulate is that late enhanced contrasts with other in vitro observations in which alteplase platelet activation may be a consequence of early reperfusion caused no change in its expression (41). occurring either directly as platelets course through the reper- When we compared the effects of reteplase and alteplase, fused myocardium or through systemically released mediators. we found few striking differences in surface antigen expression. Another implication of the present study is the failure of Indeed, from nine receptors studied, at five consecutive time platelet aggregation techniques to detect mild but significant points, the only significant differences between groups oc- changes in the platelet status in the alteplase-treated group, as curred at 24 h. At this time point, greater expression of GP revealed by flow cytometry. The temporal pattern of platelet IIb/IIIa, VLA-2 and PECAM-1 was shown in the reteplase aggregation paralleled the dynamics of receptor expression group. This finding, perhaps the most important of our study, more closely in the reteplase group, where the magnitude of suggests that adjunctive GP IIb/IIIa inhibitors may be appro- changes was greater. priately used at a later time point after thrombolysis, when Limitations of the study. There are several limitations of platelets are most active. The heightened expression of GP the present study, which was designed as a pilot investigation IIb/IIIa, VLA-2 (GP Ia/IIa [a major receptor that binds and therefore has a small sample size. Statistical differences collagen] [42]) and PECAM-1 (CD 31) (43) may play a role in between groups might be revealed in a larger cohort of rethrombosis and infarct-related vessel reocclusion. patients. The present report is primarily descriptive and does Our data correlate with the Reteplase Versus Alteplase not lead to an understanding of the mechanisms of platelet Patency Investigation During Myocardial Infarction (RAPID) activity changes after thrombolysis. Moreover, the role of most trial (31), where reteplase exhibited early patency superior to of the platelet receptors studied in the pathogenesis of AMI is that of alteplase. We speculate that the early decrease in still undefined. In addition to the observed differences in 1472 GURBEL ET AL. JACC Vol. 31, No. 7 PLATELETS DURING THROMBOLYSIS IN AMI June 1998:1466–73

We are grateful for the dependability of David R. Lowry, MPH and Helene G. Lutz, without whose assistance this study would have been impossible. We also express our appreciation to Kenneth A. Ault, MD for reviewing the ﬂow cytometry methodology, J. Richard Hebel, PhD for assistance in the statistical analysis and Barry S. Coller, MD and the GUSTO-III Steering Committee for their constructive comments and helpful suggestions on the manuscript. Patricia Williams is recognized for excellent editorial assistance.

References 1. Eccleston D, Topol EJ. Inhibitors of platelet IIb/IIIa as augmenters of thrombolysis. Coron Artery Dis 1995;6:947–55. 2. Davies MJ, Thomas AC, Knapman PA, Hangarter JR. Intramyocardial platelet aggregation in patients with acute myocardial infarction suffering sudden ischemic cardiac death. Circulation 1986;73:418–27. 3. Luchessi BR, Rote WE, Driscoll EM, Mu DX. Prevention of thrombosis and rethrombosis and enhancement of thrombolytic actions of recombinant tissue-type plasminogen activator in the canine heart by DMP728, a glycoprotein IIb/IIIa antagonist. Br J Pharmacol 1994;113:1333–43. 4. Nicolini FA, Lee P, Rios G, Kottke-Marchant K, Topol EJ. Combination of platelet fibrinogen receptor antagonist and direct thrombin inhibitor at low doses markedly improves thrombolysis. Circulation 1994;89:1802–9. 5. Roux SP, Tschopp TB, Kuhn H, Steiner B, Hadvary P. Effects of heparin, aspirin and synthetic platelet glycoprotein IIb/IIIa receptor antagonist on coronary artery reperfusion and reocclusion after thrombolysis with tissue- type plasminogen activator in the dog. J Pharmacol Exp Ther 1993;264: 501–8. 6. Yasuda T, Gold HK, Kohmura C, et al. Intravenous and endobronchial administration of G4120, a cyclic Arg-Gly-Asp-containing platelet GPIIb/ IIIa receptor-blocking pentapeptide, enhances and sustains coronary arterial thrombolysis with rt-PA in canine preparations. Arterioscl Thromb 1993;13: 738–47. 7. Ohman EM, Kleiman NS, Gacioch G, et al. Combined accelerated tissue- plasminogen activator and glycoprotein IIb/IIIa integrin receptor blockade with Integrilin in acute myocardial infarction: results of a randomized, placebo-controlled, dose-ranging trial. Circulation 1997;95:846–54. 8. Moliterno DJ, Harrington RA, Krucoff MW, et al. More complete and stable reperfusion with platelet IIb/IIIa antagonism plus thrombolysis for AMI: the PARADIGM trial [abstract]. Circulation 1996;94 Suppl I:I-553. 9. Rote WE, Mu DX, Bates ER, Nedelman MA, Lucchesi BR. Prevention of rethrombosis after coronary thrombolysis in chronic canine model: adjunc- Figure 3. Representative overlapped histograms of platelets labeled tive therapy with monoclonal antibody 7E3 F(abЈ) 2 fragment. J Cardiovasc with fluorescent antibodies to GP IIb/IIIa after reteplase (A) or Pharm 1994;23:194–202. alteplase (B) treatment. A shift to the right indicates platelet 10. Kohmura C, Gold HK, Yasuda T, et al. A chimeric murine/human antibody activation. Fab fragment directed against the platelet GP IIIa/IIb receptor enhances and sustains arterial thrombolysis with recombinant tissue-type plasminogen activator in baboons. Arterioscler Thromb Vasc Biol 1993;12:1837–42. 11. Kleiman NS, Ohman EM, Califf RM, et al. Profound inhibition of platelet reteplase and alteplase groups, clinical characteristics, such as aggregation with monoclonal antibody 7E3 Fab after thrombolytic therapy: results of the Thrombolysis and Angioplasty in Myocardial Infarction use of antecedent aspirin and time of thrombolytic agent (TAMI) 8 Pilot study. J Am Coll Cardiol 1993;22:381–9. delivery, may have influenced the results. Finally, it is possible 12. Cronberg S. Effect of fibrinolysis on adhesion and aggregation of human that we missed an early peak of platelet activation, which has platelets. Thromb Diath Haemorrh 1968;19:474–82. been reported in in vitro and animal experiments. 13. Terres W, Unmus S, Mathey DG, Bleifeld W. Effects of streptokinase, urokinase, and tissue plasminogen activator on platelet aggregability and Conclusions. In this prospective clinical ex vivo platelet stability of platelet aggregates. Cardiovasc Res 1990;67:1175–81. study, similar patterns of platelet aggregation and surface 14. Torr SR, Winters KJ, Santoro SA, Sobel BE. The nature of interactions receptor expression occurred during the first 24 h of coronary between tissue-type plasminogen activators and platelets. Thromb Res 1990;59:279–93. thrombolysis with reteplase and alteplase. However, in patients 15. Golino P, Ashton JH, Glas-Greenwalt P, McNatt J, Buja LM, Willerson JT. treated with reteplase, indicators of platelet activity were Mediation of reocclusion by thromboxane with tissue type plasminogen higher at 24 h after thrombolysis than in alteplase-treated activator in a canine preparation of coronary thrombosis. Circulation patients. These data suggest that platelet GP IIb/IIIa inhibitors 1988;77:678–84. 16. Yao SK, Ober JC, Garfinkel LI, et al. Blockade of platelet membrane or other antiplatelet strategies may be particularly advanta- glycoprotein Ib receptors delays intracoronary thrombogenesis, enhances geous when used 12 to 24 h after thrombolysis, especially after thrombolysis, and delays coronary artery reocclusion in dogs. Circulation reteplase therapy. It is at this time point during the first day of 1994;89:2822–8. 17. Fitzgerald DJ, Catella F, Roy L, Fitzgerald GA. Marked platelet activation coronary thrombolysis that GP IIb/IIIa is markedly expressed in vivo after intravenous streptokinase in patients with acute myocardial and platelets are most active. infarction. Circulation 1988;77:142–50. JACC Vol. 31, No. 7 GURBEL ET AL. 1473 June 1998:1466–73 PLATELETS DURING THROMBOLYSIS IN AMI

18. Frandsen NJ, Winther K, Pedersen F, Christiansen I, McNair P. Time course coronary thrombolysis with bolus administration of reteplase compared with of platelet ␣ granule release in acute myocardial infarction treated with alteplase infusion in acute myocardial infarction: RAPID Investigators. streptokinase. Heart 1996;75:141–4. Circulation 1995;91:2725–32. 19. Rasminis G, Vesterqvist O, Green K, Edhag O, Henriksson P. Evidence of 32. Terres W, Lund GK, Hubner A, Ehlert A, Reuter H, Mann CW. Endoge- increased platelet activation after thrombolysis in patients with acute myo- nous tissue-type plasminogen activator and platelet reactivity as risk factors cardial infarction. Br Heart J 1992;68:374–6. for reocclusion after recanalization of chronic total coronary occlusion. Am 20. Salvoni A, Marenzi G, Lauri G, et al. ␤-Thromboglobulin plasma levels in Heart J 1995;130:711–6. the first week after myocardial infarction: influence of thrombolytic therapy. 33. Coller BS. The role of platelets in arterial thrombosis and the rationale for Am Heart J 1994;128:472–6. blockade of platelet GP IIb/IIIa receptors as antithrombotic therapy. Eur 21. Bertolino G, Noris P, Previtali M, et al. Platelet function after in vivo and in Heart J 1995;16 Suppl I:11–5. vitro treatment with thrombolytic agents. Am J Cardiol 1992;69:457–61. 34. Erickson LA, Ginsberg MH, Loskutoff DJ. Detection and partial character- 22. Bihor C, Durrieu-Jais C, Besse P, Nurden P, Nurden AT. Flow cytometry istics of an inhibitor of plasminogen activator on human platelets. J Clin reveals activated GP IIb-IIIa complexes on platelets from patients undergo- Invest 1984;74:1465–72. ␣ ing thrombolytic therapy after acute myocardial infarction. Blood Coagul 35. Plow EF, Collen D. The presence and release of 2-antiplasmin from human Fibrinolysis 1995;6:395–410. platelets. Blood 1981;58:1069–74. 23. O’Brien JR, Etherington MD, Shuttleworth RD, Calwell WH. Platelet 36. Miles LA, Plow EF. Binding and activation of plasminogen on the platelet function in acute myocardial infarction patients compared to controls. surface. J Clin Invest 1986;77:2001–9. Thromb Haemost 1980;44:96–9. 37. Vaughan DE, Mendelson ME, Declerck PJ, Van Houtte E, Collen D, 24. Knudsen JB, Gormsen J, Skagen K, Amtorp O. Changes in platelet function, Loscalzo J. Characterization of the binding of human tissue-type plasmino- coagulation and fibrinolysis in uncomplicated cases of acute myocardial gen activator to platelets. J Biol Chem 1989;264:15869–74. infarction. Thromb Haemost 1979;42:1513–22. 38. Hoffmann JJML, Janssen WCM. Interactions between thrombolytic agents 25. Lu J, Qi W, Shao H. Dynamic changes of the platelet granule membrane and platelets: effects of plasmin on platelet glycoproteins Ib and IIb/IIIa. glycoprotein level in patients with acute myocardial infarction before and Thromb Res 1992;67:711–9. after thrombolysis. Chin J Intern Med 1995;34:302–4. 39. Adelman B, Michelson AD, Loscalzo J, Greenberg J, Handin RI. Plasmin 26. Coller BS. Platelets and thrombolytic therapy. N Engl J Med 1990;322:33– effect on glycoprotein Ib–von Willebrand factor interactions. Blood 1985;65: 42. 32–40. 27. The GUSTO III Investigators. A comparison of reteplase and alteplase for 40. Shafer AI, Adelman B. Plasmin inhibition of platelet function and of acute myocardial infarction. N Engl J Med 1997;337:1118–23. arachidonic acid metabolism. J Clin Invest 1985;75:456–61. 28. Ruggeri ZM. New insights into the mechanisms of platelet adhesion and 41. Kamat SG, Michelson AD, Benoit SE, et al. Fibrinolysis inhibits shear aggregation. Semin Hematol 1994;31:229–39. stress-induced platelet aggregation. Circulation 1995;92:1399–407. 29. Ault KA. Flow cytometric measurement of platelet function and reticulated 42. Staatz WD, Rajpara SM, Wayner EA, Carter WG, Santoro SA. The ϩϩ platelets. Ann N Y Acad Sci 1993;677:293–308. membrane glycoprotein Ia-IIa (VLA-2) complex mediates the Mg depen- 30. Serebruany VL, Ordonez JV, Herzog WR, et al. Dietary coenzyme Q10 dent adhesion of platelets to collagen. J Cell Biol 1989;108:1917–24. supplementation alters platelet size and inhibits human vitronectin (CD51/ 43. Newman PJ, Berndt MC, Gorsky J, White GC, Paddock LS, Muller WA. CD61) receptor expression. J Cardiovasc Pharmacol 1997;29:16–22. PECAM-1 (CD 31) cloning and relation to adhesion molecules of the 31. Smalling RW, Bode C, Kalbfleisch J et al. More rapid, complete and stable immunoglobulin gene superfamily. Science 1990;247:1219–23.