- Email: [email protected]
IIIa inhibitor
Thrombosis Research (2004) 113, 303--310
intl.elsevierhealth.com/journals/thre
Regular Article
The state of platelets preserved in extracorporeal circulation with a glycoprotein IIb/IIIa inhibitor Norihiro Kondo *, Fuminori Wakayama, Yasuyuki Suzuki, Kozo Fukui, Shunichi Takaya, Ikuo Fukuda First Department of Surgery, Hirosaki University, School of Medicine, 5 Zaifu-cho, Hirosaki, Aomori 036-8562, Japan Received 6 January 2004; received in revised form 1 March 2004; accepted 16 March 2004 Available online 06 May 2004
KEYWORDS Platelet anesthesia; Glycoprotein IIb/IIIa inhibitor; Simulated extracorporeal circulation; Platelet consumption; Flow cytometry; Amplification loop
ABSTRACT Introduction: Temporary inhibition of platelet function during extracorporeal circulation (platelet anesthesia) can preserve platelet count. We hypothesized that platelet anesthesia with a glycoprotein IIb/IIIa inhibitor could preserve activated platelets. Materials and methods: Fresh human blood from donors was recirculated for 120 min in a simulated extracorporeal circuit. Heparin and FK633, a short-acting platelet glycoprotein IIb/IIIa inhibitor, were added to recirculated blood in one group (group F, n = 5) whereas only heparin was used in controls (group C, n = 5). Blood samples were obtained from the donors, and at 0, 5, 15, 30, 60, and 120 min of recirculation. Platelet counts, beta-thromboglobulin, thrombin--antithrombin complex, and aggregation to adenosine diphosphate were measured. Flow cytometry was performed for measurement of fibrinogen binding, platelet surface expression of P-selectin, and microparticles. Results and conclusions: In the FK633 group, platelet counts were preserved and beta-thromboglobulin levels remained unchanged, whereas in group C, platelet counts decreased significantly and betathromboglobulin increased significantly from 30 and 60 min, respectively. FK633 inhibited platelet aggregation and fibrinogen binding to platelets throughout recirculation. A significant difference between groups with respect to microparticle parameters and thrombin--antithrombin complex levels was evident by 120 min. Pselectin expression increased at 0 min in both groups, and was preserved significantly at 5 min and reduced at 120 min in group F. Platelet counts were preserved by platelet anesthesia during recirculation without platelet activation.
Abbreviations: CPB, cardiopulmonarybypass; GP, glycoprotein; PE, phycoerythrin; FITC, fluorescein isothiocyanate; ADP, adenosine-5-diphosphate; PRP, platelet-rich plasma; PPP, platelet-poor plasma; EDTA, ethylenediaminetetraacetic acid; TAT, thrombin--antithrombin complex; BTG, beta-thromboglobulin. * Corresponding author. Tel.: +81-172-39-5074; fax: +81-172-37-8340. E-mail address: [email protected] (N. Kondo).
0049-3848/$ - see front matter n 2004 Elsevier Ltd. All rights reserved. doi:10.1016/j.thromres.2004.03.017
304
N. Kondo et al. / Thrombosis Research 113 (2004) 303--310 These results suggest that FK633 inhibits the amplification loop by reducing the binding of fibrinogen to glycoprotein IIb/IIIa and platelet aggregation. n 2004 Elsevier Ltd. All rights reserved.
In cardiac surgery, bleeding complications and excessive postoperative blood loss continue to contribute to t he morbidity associated with cardiopulmonary bypass (CPB) [1,2]. During CPB, blood components are exposed to artificial surfaces and platelets are activated. This is followed by the release of various inflammatory cytokines and the activation of the complement and coagulofibrinolytic systems [3], resulting in a decline in platelet number and function. However, as long as extracorporeal circulation is used, contact between artificial surfaces and the blood cannot be avoided. Heparin-coated CPB circuits [4,5] and new materials mimicking endothelial cells have thus been developed [6,7]. Although these circuits reduce systemic inflammatory reactions and lower platelet activation, platelets can also be activated by other factors, such as aspirated blood collected from mediastinal cavities and/or pleural cavities [7], and heparin administration [8]. Platelet anesthesia is another strategy for preserving platelet number and function, involving temporary pharmacologic inhibition of platelets during extracorporeal circulation [9,10]. Platelet aggregation is caused by activation of glycoprotein (GP) IIb/IIIa complex and cross-linking with fibrinogen [11]. We hypothesized that platelet anesthesia with FK633, a shortacting glycoprotein IIb/IIIa inhibitor, could preserve activated platelets due to inhibition of the final common pathway of platelet aggregation. In order to develop a strategy for platelet anesthesia, it is important to clarify the functional state in which platelets can be preserved. We used whole-blood flow cytometry for direct analysis of individual platelets [12,13] and measured plasma levels of soluble markers reflecting platelet activation for additional indication of platelet activity.
Materials and methods Reagents FK633 (N-{4-(4-Amidinophenoxy)butyryl}-alpha-Laspartyl-L-valine) was obtained from Fujisawa Pharmaceutical (Osaka, Japan). Phycoerythrin (PE)-conjugated anti-GPIb antibody for platelet and microparticle detection, and fluorescein isothiocyanate (FITC)-conjugated anti-P-selectin antibody were obtained from Pharmingen (San Diego, CA).
FITC-labeled affinity-purified chicken anti-fibrinogen antibodies were obtained from Immunsystem (Uppsala, Sweden). Adenosine-5-diphosphate (ADP) was supplied by MC Medical (Tokyo, Japan) and heparin sodium by Aventis Pharma Japan (Tokyo, Japan).
In vitro platelet aggregation study at several concentrations of FK633 Blood was withdrawn from healthy volunteers (n = 6) into tubes containing 1/10 volume of 3.8% sodium citrate and then centrifuged at 100 g for 10 min before collection of platelet-rich plasma (PRP). The remaining blood was centrifuged for a further 10 min at 1500 g to prepare platelet-poor plasma (PPP). The aggregation assay was performed using a laser light scattering aggregometer (PA-200, Kowa, Tsukuba, Japan), with transmittance of PPP being calibrated as 100%. PRP (240 Al) was incubated in the aggregometer with FK633 (30 Al, at final concentrations of 100 nmol/l, 300 nmol/l, 1 mmol/l, and 3 mmol/l) for 2 min at 37 jC. ADP (30 Al) was then added, of which the final concentration was 5 Amol/l. Light transmittance was recorded and percent inhibition was calculated from control aggregation. FK633 dosage was determined based on the concentration of submaximum inhibition obtained from dose--response data.
Simulated extracorporeal circulation The simulated extracorporeal circuit incorporated silicon rubber tubing (1/4- and 3/8-in. internal diameter, MERA, Tokyo, Japan), a polyvinyl chloride reservoir bag (MERA), a spiral coil membrane oxygenator (surface area 1.0 m2, MERA), and an occlusive roller head pump (MERA). Each circuit was primed with 200 ml of fresh human blood obtained from healthy volunteers (n = 10) who were unmedicated at least for 14 days before donation. Donors were prospectively divided into two groups. In one group (group F, n = 5), heparin sodium (5 units/ml blood) and FK633, a short-acting platelet glycoprotein IIb/IIIa inhibitor (0.001 mg/ml blood), were added to a reservoir bag containing 150 ml lactated Ringer’s solution and glucose (3.75 mg/ml blood) prior to circulation. In controls (group C, n = 5), only heparin (5 units/ml blood) was added. Blood was drawn directly into the reservoir bag and was
N. Kondo et al. / Thrombosis Research 113 (2004) 303--310 recirculated for 120 min at 1000 ml/min with the reservoir bag kept in a constant temperature water bath to maintain the temperature at 37 jC. The oxygenator was ventilated with 100% oxygen at a rate of 1.0 l/min. Blood samples were obtained for analysis from each donor (baseline), from the reservoir bag before beginning recirculation (time point 0), and at 5, 15, 30, 60, and 120 min of recirculation.
Blood samples and assays Blood samples were obtained with ethylenediaminetetraacetic acid (EDTA)--2Na 1.5 mg/ml blood for cell counts, and with 3.8% sodium citrate for platelet aggregation and flow cytometry. Blood counts were measured using a fully automated blood cell analyzer (SF-3000 Sysmex, Kobe, Japan), while platelet aggregations were measured as described above. At each sampling point, blood was withdrawn for measurement of plasma levels of soluble markers reflecting platelet activation. These samples were divided into collection tubes containing theophylline, adenosine, dipyridamole, and sodium citrate for determination of beta-thromboglobulin (BTG), 3.8% sodium citrate for thrombin--antithrombin complex (TAT) and fibrinogen assays, or trasyrol, trypsin inhibitor, and protamine sulfate for bradykinin assay. Plasma samples were separated by centrifugation at 2000 g for 10 min at 4 jC after 15 min cooling with ice. These markers were measured at SRL (Tokyo, Japan) and all assay procedures followed the manufacturer’s protocols. For flow cytometry, 5 Al whole blood was added to polystyrene tubes containing 100 Al HEPES buffer (137 mmol/l NaCl, 2.7 mmol/l KCl, 1 mmol/ l MgCl2, 5.6 mmol/l glucose, 1 g/l bovine serum albumin, and 20 mmol/l HEPES, pH 7.4) and 10 Al PE-conjugated anti-GPIb antibody and 10 Al FITC-conjugated antibody. Samples were incubated for 10 min at room temperature and were then diluted and fixed with 1000 Al ice-cold phosphate-buffered saline (0.02 mol/l Na2HPO4, 0.15 mol/l NaCl, 0.02% NaN3, pH 7.2), containing 1% pformaldehyde. No steps that induce platelet activation such washing, centrifugation, vortexing, or stirring were used. Flow cytometry was performed utilizing FACScan (Becton Dickinson, Tokyo, Japan). Data processing from 10000 platelets was carried out with the Cell Quest (Becton Dickinson), an instrument that characterizes PE fluorescencepositive (GP Ib-positive) cells as platelets. Based on its light scattering properties, each cell is represented by a point in a rectangular coordinate system and a discrimination frame is placed
305
around the platelet cluster. The instrument measures the complexity (right angle scatter), cell size (forward angle scatter), and percentage of platelets demonstrating surface expression of P-selectin and fibrinogen binding. Analytical markers were set in the forward angle scatter channel to divide the negative control sample into two fractions; one containing 98--99% of the platelets, and the other containing the smallest platelets, 1--2% of the total. Those platelets with forward scatter lower than the marker were identified as microparticles. Results were expressed as the percentage of anti-GP Ib antibody-positive platelets, defined as P-selectin- or fibrinogen-positive with a fluorescence intensity exceeding that of 98--99% of control platelets.
Statistical analysis All data are reported as mean F standard deviation. We performed analyses using StatView for Windows J5.0 (SAS institute, North Carolina). The two groups were compared using the nonparametric Mann--Whitney U-test at each time point and differences were considered to be statistically significant when the p-value was less than 0.05.
Results Inhibition of platelet aggregation to ADP by FK633 Inhibition of platelet aggregation was 59 F 10%, 76 F 8.2%, 96 F 3.8%, and 99 F 2.8% at final concentrations of 100 nmol/l, 300 nmol/l, 1 mmol/l, and 3 mmol/l, respectively. Platelet aggregation in PRP was thus inhibited dose-dependently. We therefore decided to administer a dosage of 0.2 mg FK633 to the circuit to give a final concentration of 1 mmol/l.
Blood counts At baseline, no significant differences in differential blood count were observed between the two groups: white blood cells (group F; 4.36 F 0.49 vs. group C; 5.64 F 1.43, p = 0.14), red blood cells (F; 4.63 F 0.1 vs. C; 4.83 F 0.38, p = 0.58), hemoglobin (F; 14.6 F 1.2 vs. C; 15.1 F 1.6, p = 0.67), hematocrit (F; 41.9 F 3.2 vs. C; 43.9 F 4.4, p = 0.49), and platelet counts (F; 193 F 43.9 vs. C; 226 F 52.9, p = 0.31). All values but platelet counts did not change significantly throughout
306
Measures of platelet function at each time point
Variable Platelet count (% of baseline) Platelet aggregation (% of baseline) Beta-thromboglobulin (ng/ml) Thrombin--antithrombin complex (Ag/ml) Bradykinin (pg/ml) Fibrinogen (mg/dl) Fibrinogen binding (% of positive platelets) P-selectin (% of positive platelets) Microparticles (/10000 GP Ib-positive cells)
Group F C F C F C F C F C F C F C F C F C
Baseline 100 100 100 100 464 F 159 317 F 190 2.3 F 1.0 3.7 F 2.0 22.5 F 12.7 32.5 F 2.9 234 F 81 175 F 15 2.7 F 0.4 2.4 F 0.5 0.4 F 0.2 0.4 F 0.3 0.19 F 0.12 0.10 F 0.10
Bag 98.9 F 11.7 90.3 F 12.9 18.7 F 7.4 54.2 F 17.6** 1623 F 1209 1678 F 1358 3.9 F 2.4 2.9 F 0.9 30.2 F 21.8 41.2 F 5.6 158 F 47 122 F 8 3.5 F 1.1 10.0 F 4.8** 30.7 F 7.0 26.3 F 6.8 0.15 F 0.10 0.14 F 0.15
Recirculation time (min) 5
15
30
60
120
93.3 F 10.2 83.4 F 4.6 22.1 F 8.4 71.5 F 25.3** 1663 F 1251 2250 F 1118 3.8 F 1.9 3.3 F 1.2 32.0 F 24.7 40.1 F 9.1 145 F 63 122 F 13 3.4 F 1.6 6.7 F 3.4 28.3 F 3.5 17.0 F 4.9* 0.42 F 0.3 0.13 F 0.1
88.8 F 11.0 76.2 F 4.4 27.5 F 13.1 64.4 F 19.5** 1642 F 1211 2498 F 1049 3.1 F 3.1 4.5 F 2.9 14.1 F 12.3 36.9 F 6.7 148 F 60 121 F 12 2.9 F 0.9 5.5 F 2.0 24.4 F 5.8 17.3 F 5.2* 0.28 F 0.18 0.13 F 0.12
86.8 F 12.4 59.8 F 13.4* 18.0 F 10.7 68.6 F 25.9** 1708 F 1285 2862 F 1134 4.4 F 2.3 5.6 F 4.3 18.0 F 10.8 37.0 F 7.8 143 F 52 125 F 9 2.5 F 1.0 4.6 F 1.9 20.1 F 6.9 19.3 F 3.7 0.30 F 0.26 0.15 F 0.06
83.4 F 13.2 43.6 F 15.8** 21.0 F 11.6 79.8 F 25.7** 1615 F 1134 3952 F 1129* 3.6 F 1.7 7.0 F 4.7 15.6 F 11.5 35.3 F 9.2 154 F 50 124 F 13 2.9 F 0.8 8.1 F 4.5* 16.0 F 5.2 12.8 F 3.4 0.28 F 0.17 0.58 F 0.39
77.4 F 8.7 35.1 F 12.1** 23.5 F 11.1 54.4 F 17.7** 1986 F 1147 5629 F 1711* 6.4 F 1.8 15.6 F 5.7* 15.9 F 8.7 33.5 F 8.5 159 F 56 124 F 7 2.7 F 0.9 11.0 F 4.7** 5.8 F 5.0 15.3 F 5.8* 0.34 F 0.20 3.25 F 3.03
Data are expressed as mean and standard deviation of mean. * p < 0.05 between the control group (C) and the group given FK633 (F) (Mann--Whitney U-test). ** p < 0.01 between the control group (C) and the group given FK633 (F) (Mann--Whitney U-test).
N. Kondo et al. / Thrombosis Research 113 (2004) 303--310
Table 1
N. Kondo et al. / Thrombosis Research 113 (2004) 303--310 recirculation except due to dilution with the circuit contents (data not shown). However, in both groups, significant changes in platelet count oc-
307
curred over time (Table 1; Fig. 1A). In group C, platelet counts after 30-min circulation were significantly lower than those at baseline and signif-
Fig. 1 Effects of FK633 on (A) platelet counts, (B) beta-thromboglobulin, (C) fibrinogen binding, (D) P-selectin expression and (E) microparticle. Data are expressed as triangle and continuous line (group F, recirculated with heparin and FK633) and rhombs and dotted line (group C, recirculated with heparin). Data points represent mean, error bars represent standard deviation. *p < 0.05; **p < 0.01 (comparison using the nonparametric Mann--Whitney U-test at each time points).
308
N. Kondo et al. / Thrombosis Research 113 (2004) 303--310
icantly lower than those in group F at the same sample time.
Platelet aggregation to ADP (Table 1) Platelet aggregation was significantly lower in group F at time point 0 and remained at the same level until the end of recirculation.
Soluble markers (Table 1) In both groups, levels of BTG increased at time point 0. BTG levels in group F remained unchanged thereafter, whereas levels in group C increased with ongoing recirculation (Fig. 1B). TAT levels significantly increased at 120 min after simulated extracorporeal circulation in group C, but no increase was observed in group F. In both groups, levels of fibrinogen decreased at time point 0 and remained unchanged throughout recirculation. In both groups, bradykinin levels increased at time point 0 and decreased with recirculation. However, changes in bradykinin levels did not reach significance.
Flow cytometry In group F, percentage of platelets exhibiting fibrinogen binding remained unchanged throughout recirculation, whereas that in group C increased significantly at time point 0. In group C, platelet fibrinogen binding decreased with recirculation, but increased at 60 and 120 min (Fig. 1C). In both groups, P-selectin expression increased at time point 0. Although P-selectin expression was preserved in group F at 5 min of recirculation, it reduced consistently over time, and fell to levels below those seen in group C by 120 min (Fig. 1D). The percentage of microparticles increased 120 min after simulated extracorporeal circulation in group C (Fig. 1E); however, no increase in microparticles was observed in group F.
Discussion Bleeding complications are a constant problem after cardiac surgery, resulting in 4% of patients requiring reentry into the operating theater after the first 24 h post surgery [1]. Such complications are related to the use of heparin, the onset of fibrinolysis, platelet loss and dysfunction, and a combination of these factors [14]. Platelets are partially activated by heparin and are diluted, destroyed, and modified during CPB because of adhesion to circuit surfaces,
aggregation, and activation [2]. Therefore, after CPB, a reduction in platelet number and function prolongs bleeding times and contributes to postoperative bleeding. Argatroban, a short-acting direct thrombin inhibitor, was used in a patient with heparin induced thrombocytopenia and was found not to have a platelet-activating effect. However, unlike heparin, which can be neutralized with protamine, argatroban has no neutralizing agent [15] and it is therefore impossible to reverse its action at the end of extracorporeal circulation. Therefore, heparin could not currently be replaced by argatroban. In another study, platelets that had become microaggregated due to heparin administration during extracorporeal circulation were preferentially lost from circulation when compared to single platelets [8]. Nonetheless, there is a possibility that, despite activation, platelets can be preserved by temporarily inhibiting their aggregation. Glycoprotein IIb/IIIa inhibitors target the final common pathway of platelet aggregation, which involves the cross-linking of glycoprotein IIb/IIIa complex, a receptor for adhesive plasma proteins, particularly fibrinogen. Previous studies have validated the concept of platelet anesthesia using a GP IIb/IIIa inhibitor to prevent loss of platelet number and function during CPB [9,10,16]. We hypothesized that platelet anesthesia with FK633, a short-acting GP IIb/IIIa inhibitor, functions by keeping activated platelets single. This study revealed that platelet anesthesia with FK633 could preserve platelet numbers to an extent similar to that reported previously [17,18] using other GP IIb/IIIa inhibitors. From the results of BTG levels in the present study, platelets were already activated when blood was drawn from donor to reservoir bag. BTG levels correlated closely with platelet counts, rising during recirculation in group C, while platelet count simultaneously decreased. Difference of BTG levels between the two groups was significant from 60 min, but evident from 5-min recirculation. Mollnes et al. [20] reported that platelet counts and BTG quantification could be used to reliably evaluate the degree of platelet biocompatibility. Although platelet counts are undoubtedly important, as mentioned in the study above, they provide no direct indication of platelet activation. Platelet counts are useful if combined with methods directly reflecting activation [20]. To ascertain the effects of FK633, we performed platelet aggregation and flow cytometry for direct analysis of individual platelets, in a manner similar to that of Mollnes et al. We confirmed our hypothesis based on these results. Results of platelet aggregation to ADP and from flow cytometry of fibrinogen binding indicated that FK633 preserves platelets by inhibition of
N. Kondo et al. / Thrombosis Research 113 (2004) 303--310 fibrinogen binding to GP IIb/IIIa complex. Effects of varying FK633 dose on platelet aggregation were comparable with previous reports [19]. The percentage inhibition in group F actually remained at about 80% although we administered FK633 so as to inhibit submaximally on the basis of the current and previous studies. Musial et al. [18] reported that short disintegrins have a higher affinity for resting platelets than do medium disintegrins. FK633 may also have higher affinity for resting platelets. It is thought that the action of FK633 declined because platelets had already become activated while in the reservoir bag. In group F, fibrinogen did not bind to platelets, whereas in group C fibrinogen binding was seen at time point 0 and decreased rapidly with recirculation. This finding suggested that platelets cross-linked with fibrinogen were consumed from the beginning of recirculation. It appears that the increase in proportion of fibrinogen-positive platelets is based on reduction of platelets due to consumption. P-selectin-positive platelets increased in parallel with BTG levels in both groups at time point 0. These flow cytometry results support the notion that platelets exposed to FK633 exist in the extracorporeal circuit in the activated single state, and that as a result, rapid platelet consumption did not occur at initiation of recirculation. However, platelet consumption did occur over time in group F, suggesting that the activated GP IIb/IIIa complex might cause fibrinogen binding and aggregation during recirculation. In actual cardiac surgery, administration of heparin occurs before initiation of cardiopulmonary bypass. Administration of FK633 before heparin might therefore suppress the activation of platelets by the circuit and by heparin. BTG levels and P-selectin expression indicated that platelet activation during extracorporeal circulation did not occur in group F. FK633 is thought to inhibit an amplification loop in platelet aggregation, involving release of molecules such as ADP and 5hydroxytryptamine (serotonin) from activated platelets, after fibrinogen binding [21]. We had hypothesized that P-selectin-positive platelets would increase during recirculation due to preservation of activated platelets. However, based on these results, the effects of FK633 on platelet preservation result from the inhibition of platelet activation during recirculation. Levels of TAT complex, an indication of plasma thrombin level induced by an activated coagulation system, increased in group C in which only heparin was administered. In group F, when a highly specific antiplatelet agent [19] was added, TAT levels were unchanged, suggesting that thrombin production might also
309
contribute to activation of platelets during simulated ECC. These findings are in accordance with the results obtained by Carroll et al. [22] and by Li et al. [23], which showed inhibition of a-granule release and strong inhibition of thrombin generation with GP IIb/IIIa inhibitors. Bradykinin reflects activation of the contact system. Contact with the artificial surfaces of the extracorporeal circuit activates the intrinsic coagulation pathway in which bradykinin is produced by the kallikrein-kinin system. Differences in bradykinin levels between the two groups did not reach significance, suggesting that the simulated extracorporeal circuit had little influence on the intrinsic coagulation pathway. Plasma levels of fibrinogen tend to decrease as fibrinogen is deposited within the circuit. Since fibrinogen levels remained unchanged despite platelet consumption, the amount of plasma fibrinogen appeared to be more than sufficient. In the present study, the simulated extracorporeal circuit, an entirely in-vitro system, conferred the advantage that results reflected platelet activation due to recirculation alone, because new platelets could not be recruited from the bone marrow, liver, or spleen. In addition, since this system avoided the effects of hepatic or renal metabolism and excretion on soluble markers of platelet activation, levels of markers were considered to reflect the difference between the two groups more clearly. Though further confirmation is needed for the application of this in-vitro study to in-vivo conditions, there are several reports of the perioperative use of short-acting GP IIb/IIIa inhibitors in cardiopulmonary bypass in clinical or animal experiments. Gammie et al. [24] reported that abciximab, when given within 12 h of cardiac surgery, was associated with significantly increased blood loss and transfusion. Abciximab is the Fab fragment of chimeric mouse/human immunoglobulin G antibody. Although its plasma half-life is 26 min, its antiplatelet effect is prolonged and irreversible after administration [25,26]. Bizzarri et al. [1] reported that the infusion of tirofiban hydrochloride immediately before surgical intervention did not adversely affect clinical outcome and had no deleterious effect on postoperative bleeding, maintaining hemoglobin and platelet counts at stable levels. Suzuki et al. [9,10] reported that eptifibatide reduces platelet activation during cardiopulmonary bypass and accelerates restoration of normal bleeding times after operation in a baboon model. The effects of tirofiban and eptifibatide are reversible. On the other hand, FK633 has a reversible antiplatelet effect and a plasma half-life of 0.52 h [19], which
310
N. Kondo et al. / Thrombosis Research 113 (2004) 303--310
is shorter than those of tirofiban and eptifibatide. Therefore FK633 may be able to protect platelets and decrease postoperative bleeding. In conclusion, platelet counts could be maintained by platelet anesthesia with FK633 during extracorporeal circulation. Although FK633 could not completely prevent consumption of platelets, platelets were not activated despite prolonged recirculation. These results suggest that FK633 inhibits the amplification loop by inhibiting the binding of fibrinogen to glycoprotein IIb/IIIa and platelet aggregation.
Acknowledgements The authors thank Ms. Yuko Tsushima for flow cytometry, Mr. Masakatsu Nakagawara (MERA) for creation of perfusion circuits and supervision of recirculation, and Mr. Yuta Inoue and Ms. Maya Ohashi for their excellent technical support. We also are indebted to Dr. Yoshiko Tamai for helpful discussions of the manuscript.
References [1] Bizzarri F, Scolletta S, Tucci E, Lucidi M, Davoli G, Toscano T, et al. Perioperative use of tirofiban hydrochloride (Aggrastat) does not increase surgical bleeding after emergency or urgent coronary artery bypass grafting. J Thorac Cardiovasc Surg 2001;122:1181--5. [2] Woodman RC, Harker LA. Bleeding complication associated with cardiopulmonary bypass. Blood 1990;76:1680--97. [3] Tanaka K, Takao M, Yada I, Yuasa H, Kugasawa M, Deguchi K. Alteration in coagulation and fibrinolysis associated with cardiopulmonary bypass during open heart surgery. J Cardiovasc Anesth 1989;3:181--8. [4] Gu YJ, van Oeveren W, Akkerman C, Boonstra PW, Huyzen RJ, Wildevuur CR. Heparin-coated circuits reduce the inflammatory response to cardiopulmonary bypass. Ann Thorac Surg 1993;55:917--22. [5] Johnell M, Elgue G, Larsson R, Larsson A, Thelin S, Siegbahn A. Coagulation, fibrinolysis, and cell activation in patients and shed mediastinal blood during coronary artery bypass grafting with a new heparin-coated surface. J Thorac Cardiovasc Surg 2002 August;124(2):321--32. [6] De Somer F, Francßois K, van Oeveren W, Poelaert J, De Wolf D, Ebels T, et al. Phosphorylcholine coating of extracorporeal circuits provides natural protection against blood activation by the material surface. Eur J Cardiothorac Surg 2000;18:602--6. [7] De Somer F, Van Belleghem Y, Caes F, Francßois K, Van Overbeke H, Arnout J, et al. Tissue factor as the main activator of the coagulation system during cardiopulmonary bypass. J Thorac Cardiovasc Surg 2002;123:951--8. [8] Muriithi EW, Belcher PR, Rao JN, Chaudhry MA, Nicol D, Wheatley DJ. The effects of heparin and extracorporeal circulation on platelet counts and platelet microaggregation during cardiopulmonary bypass. J Thorac Cardiovasc Surg 2000;120:538--43.
[9] Suzuki Y, Hillyer P, Miyamoto S, Niewiarowski S, Sun L, Rao AK, et al. Integrilin prevents prolonged bleeding times after cardiopulmonary bypass. Ann Thorac Surg 1998;66: 373--81. [10] Suzuki Y, Malekan R, WilliamHanson CW, Niewiarowski S, Sun L, Rao AK, et al. Platelet anesthesia with nitric oxide with or without eptifibatide during cardiopulmonary bypass in baboons. J Thorac Cardiovasc Surg 1999;117:987--93. [11] Ruoslahti E. Integrins. J Clin Invest 1991;87:1--5. [12] Yamazaki M, Uchiyama S, Iwata M. Measurement of platelet fibrinogen binding and P-selectin expression by flow cytometry in patients with cerebral infarction. Thromb Res 2001;104:197--205. [13] Michelson AD. Flow cytometry: a clinical test of platelet function. Blood 1996;87:4925--36. [14] Khuri SF, Valeri CR, Loscalzo J, Weinstein MJ, Birjiniuk V, Healey NA, et al. Heparin causes platelet dysfunction and induces fibrinolysis before cardiopulmonary bypass. Ann Thorac Surg 1995;60:1008--14. [15] Levy JH. Pharmacologic preservation of the hemostatic system during cardiac surgery. Ann Thorac Surg 2001;72: S1814--20. [16] Hiramatsu Y, Gikakis N, Anderson III HL, Gorman III JH, Marcinkiewicz C, Gould RJ, et al. Tirofiban provides ‘‘platelet anesthesia’’ during cardiopulmonary bypass in baboons. J Thorac Cardiovasc Surg 1997;113:182--93. [17] Bernabei A, Gikakis N, Kowalska MA, Niewiarowski S, Edmunds Jr LH. Iloprost and echistatin protect platelets during simulated extracorporeal circulation. Ann Thorac Surg 1995;59:149--53. [18] Musial J, Niewiarowski S, Rucinski B, Stewart GJ, Cook JJ, Williams JA, et al. Inhibition of platelet adhesion to surfaces of extracorporeal circuits by disintegrins. RGD-containing peptides from viper venoms. Circulation 1990;82(1): 261--73. [19] Aoki T, Cox D, Senzaki K, Seki J, Tanaka A, Takasugi H, et al. The anti-platelet and anti-thrombotic effects of FK633, a peptide-mimetic GPIIB/IIIA antagonist. Thromb Res 1996;81:439--50. [20] Mollnes TE, Videm V, Christiansen D, Bergseth G, Riesenfeld J, Hovig T. Platelet compatibility of an artificial surface modified with functionally active heparin. Thromb Haemost 1999;82:1132--6. [21] Hong J, Larsson A, Ekdahl KN, Elgue G, Larsson R, Nilsson B. Contact between a polymer and whole blood: sequence of events leading to thrombin generation. J Lab Clin Med 2001;138:139--45. [22] Carroll RC, Wang XF, Lanza F, Steiner B, Kouns WC. Blocking platelet aggregation inhibits thromboxane A2 formation by low dose agonists but does not inhibit phosphorylation and activation of cytosolic phospholipase A2. Thromb Res 1997;88:109--25. [23] Li N, Wallen NH, Savi P, Herault JP, Herbert JM. Effects of a new platelet glycoprotein IIb/IIIa antagonist, SR121566, on platelet activation, platelet--leukocyte interaction and thrombin generation. Blood Coagul Fibrinolysis 1998;9: 507--15. [24] Gammie JS, Zenati M, Kormos RL, Hattler BG, Wei LM, Pellegrini RV, et al. Abciximab and the risk of bleeding during emergency cardiac operations. Ann Thorac Surg 1999; 67:292--3. [25] Coller BS. GPIIb/IIIa antagonists: pathophysiologic and therapeutic insights from studies of c7E3 Fab. Thromb Haemost 1997;78:730--5. [26] Kleiman NS. Pharmacokinetics and pharmacodynamics of glycoprotein IIb--IIIa inhibitors. Am Heart J 1999;138: 263--75.
intl.elsevierhealth.com/journals/thre
Regular Article
The state of platelets preserved in extracorporeal circulation with a glycoprotein IIb/IIIa inhibitor Norihiro Kondo *, Fuminori Wakayama, Yasuyuki Suzuki, Kozo Fukui, Shunichi Takaya, Ikuo Fukuda First Department of Surgery, Hirosaki University, School of Medicine, 5 Zaifu-cho, Hirosaki, Aomori 036-8562, Japan Received 6 January 2004; received in revised form 1 March 2004; accepted 16 March 2004 Available online 06 May 2004
KEYWORDS Platelet anesthesia; Glycoprotein IIb/IIIa inhibitor; Simulated extracorporeal circulation; Platelet consumption; Flow cytometry; Amplification loop
ABSTRACT Introduction: Temporary inhibition of platelet function during extracorporeal circulation (platelet anesthesia) can preserve platelet count. We hypothesized that platelet anesthesia with a glycoprotein IIb/IIIa inhibitor could preserve activated platelets. Materials and methods: Fresh human blood from donors was recirculated for 120 min in a simulated extracorporeal circuit. Heparin and FK633, a short-acting platelet glycoprotein IIb/IIIa inhibitor, were added to recirculated blood in one group (group F, n = 5) whereas only heparin was used in controls (group C, n = 5). Blood samples were obtained from the donors, and at 0, 5, 15, 30, 60, and 120 min of recirculation. Platelet counts, beta-thromboglobulin, thrombin--antithrombin complex, and aggregation to adenosine diphosphate were measured. Flow cytometry was performed for measurement of fibrinogen binding, platelet surface expression of P-selectin, and microparticles. Results and conclusions: In the FK633 group, platelet counts were preserved and beta-thromboglobulin levels remained unchanged, whereas in group C, platelet counts decreased significantly and betathromboglobulin increased significantly from 30 and 60 min, respectively. FK633 inhibited platelet aggregation and fibrinogen binding to platelets throughout recirculation. A significant difference between groups with respect to microparticle parameters and thrombin--antithrombin complex levels was evident by 120 min. Pselectin expression increased at 0 min in both groups, and was preserved significantly at 5 min and reduced at 120 min in group F. Platelet counts were preserved by platelet anesthesia during recirculation without platelet activation.
Abbreviations: CPB, cardiopulmonarybypass; GP, glycoprotein; PE, phycoerythrin; FITC, fluorescein isothiocyanate; ADP, adenosine-5-diphosphate; PRP, platelet-rich plasma; PPP, platelet-poor plasma; EDTA, ethylenediaminetetraacetic acid; TAT, thrombin--antithrombin complex; BTG, beta-thromboglobulin. * Corresponding author. Tel.: +81-172-39-5074; fax: +81-172-37-8340. E-mail address: [email protected] (N. Kondo).
0049-3848/$ - see front matter n 2004 Elsevier Ltd. All rights reserved. doi:10.1016/j.thromres.2004.03.017
304
N. Kondo et al. / Thrombosis Research 113 (2004) 303--310 These results suggest that FK633 inhibits the amplification loop by reducing the binding of fibrinogen to glycoprotein IIb/IIIa and platelet aggregation. n 2004 Elsevier Ltd. All rights reserved.
In cardiac surgery, bleeding complications and excessive postoperative blood loss continue to contribute to t he morbidity associated with cardiopulmonary bypass (CPB) [1,2]. During CPB, blood components are exposed to artificial surfaces and platelets are activated. This is followed by the release of various inflammatory cytokines and the activation of the complement and coagulofibrinolytic systems [3], resulting in a decline in platelet number and function. However, as long as extracorporeal circulation is used, contact between artificial surfaces and the blood cannot be avoided. Heparin-coated CPB circuits [4,5] and new materials mimicking endothelial cells have thus been developed [6,7]. Although these circuits reduce systemic inflammatory reactions and lower platelet activation, platelets can also be activated by other factors, such as aspirated blood collected from mediastinal cavities and/or pleural cavities [7], and heparin administration [8]. Platelet anesthesia is another strategy for preserving platelet number and function, involving temporary pharmacologic inhibition of platelets during extracorporeal circulation [9,10]. Platelet aggregation is caused by activation of glycoprotein (GP) IIb/IIIa complex and cross-linking with fibrinogen [11]. We hypothesized that platelet anesthesia with FK633, a shortacting glycoprotein IIb/IIIa inhibitor, could preserve activated platelets due to inhibition of the final common pathway of platelet aggregation. In order to develop a strategy for platelet anesthesia, it is important to clarify the functional state in which platelets can be preserved. We used whole-blood flow cytometry for direct analysis of individual platelets [12,13] and measured plasma levels of soluble markers reflecting platelet activation for additional indication of platelet activity.
Materials and methods Reagents FK633 (N-{4-(4-Amidinophenoxy)butyryl}-alpha-Laspartyl-L-valine) was obtained from Fujisawa Pharmaceutical (Osaka, Japan). Phycoerythrin (PE)-conjugated anti-GPIb antibody for platelet and microparticle detection, and fluorescein isothiocyanate (FITC)-conjugated anti-P-selectin antibody were obtained from Pharmingen (San Diego, CA).
FITC-labeled affinity-purified chicken anti-fibrinogen antibodies were obtained from Immunsystem (Uppsala, Sweden). Adenosine-5-diphosphate (ADP) was supplied by MC Medical (Tokyo, Japan) and heparin sodium by Aventis Pharma Japan (Tokyo, Japan).
In vitro platelet aggregation study at several concentrations of FK633 Blood was withdrawn from healthy volunteers (n = 6) into tubes containing 1/10 volume of 3.8% sodium citrate and then centrifuged at 100 g for 10 min before collection of platelet-rich plasma (PRP). The remaining blood was centrifuged for a further 10 min at 1500 g to prepare platelet-poor plasma (PPP). The aggregation assay was performed using a laser light scattering aggregometer (PA-200, Kowa, Tsukuba, Japan), with transmittance of PPP being calibrated as 100%. PRP (240 Al) was incubated in the aggregometer with FK633 (30 Al, at final concentrations of 100 nmol/l, 300 nmol/l, 1 mmol/l, and 3 mmol/l) for 2 min at 37 jC. ADP (30 Al) was then added, of which the final concentration was 5 Amol/l. Light transmittance was recorded and percent inhibition was calculated from control aggregation. FK633 dosage was determined based on the concentration of submaximum inhibition obtained from dose--response data.
Simulated extracorporeal circulation The simulated extracorporeal circuit incorporated silicon rubber tubing (1/4- and 3/8-in. internal diameter, MERA, Tokyo, Japan), a polyvinyl chloride reservoir bag (MERA), a spiral coil membrane oxygenator (surface area 1.0 m2, MERA), and an occlusive roller head pump (MERA). Each circuit was primed with 200 ml of fresh human blood obtained from healthy volunteers (n = 10) who were unmedicated at least for 14 days before donation. Donors were prospectively divided into two groups. In one group (group F, n = 5), heparin sodium (5 units/ml blood) and FK633, a short-acting platelet glycoprotein IIb/IIIa inhibitor (0.001 mg/ml blood), were added to a reservoir bag containing 150 ml lactated Ringer’s solution and glucose (3.75 mg/ml blood) prior to circulation. In controls (group C, n = 5), only heparin (5 units/ml blood) was added. Blood was drawn directly into the reservoir bag and was
N. Kondo et al. / Thrombosis Research 113 (2004) 303--310 recirculated for 120 min at 1000 ml/min with the reservoir bag kept in a constant temperature water bath to maintain the temperature at 37 jC. The oxygenator was ventilated with 100% oxygen at a rate of 1.0 l/min. Blood samples were obtained for analysis from each donor (baseline), from the reservoir bag before beginning recirculation (time point 0), and at 5, 15, 30, 60, and 120 min of recirculation.
Blood samples and assays Blood samples were obtained with ethylenediaminetetraacetic acid (EDTA)--2Na 1.5 mg/ml blood for cell counts, and with 3.8% sodium citrate for platelet aggregation and flow cytometry. Blood counts were measured using a fully automated blood cell analyzer (SF-3000 Sysmex, Kobe, Japan), while platelet aggregations were measured as described above. At each sampling point, blood was withdrawn for measurement of plasma levels of soluble markers reflecting platelet activation. These samples were divided into collection tubes containing theophylline, adenosine, dipyridamole, and sodium citrate for determination of beta-thromboglobulin (BTG), 3.8% sodium citrate for thrombin--antithrombin complex (TAT) and fibrinogen assays, or trasyrol, trypsin inhibitor, and protamine sulfate for bradykinin assay. Plasma samples were separated by centrifugation at 2000 g for 10 min at 4 jC after 15 min cooling with ice. These markers were measured at SRL (Tokyo, Japan) and all assay procedures followed the manufacturer’s protocols. For flow cytometry, 5 Al whole blood was added to polystyrene tubes containing 100 Al HEPES buffer (137 mmol/l NaCl, 2.7 mmol/l KCl, 1 mmol/ l MgCl2, 5.6 mmol/l glucose, 1 g/l bovine serum albumin, and 20 mmol/l HEPES, pH 7.4) and 10 Al PE-conjugated anti-GPIb antibody and 10 Al FITC-conjugated antibody. Samples were incubated for 10 min at room temperature and were then diluted and fixed with 1000 Al ice-cold phosphate-buffered saline (0.02 mol/l Na2HPO4, 0.15 mol/l NaCl, 0.02% NaN3, pH 7.2), containing 1% pformaldehyde. No steps that induce platelet activation such washing, centrifugation, vortexing, or stirring were used. Flow cytometry was performed utilizing FACScan (Becton Dickinson, Tokyo, Japan). Data processing from 10000 platelets was carried out with the Cell Quest (Becton Dickinson), an instrument that characterizes PE fluorescencepositive (GP Ib-positive) cells as platelets. Based on its light scattering properties, each cell is represented by a point in a rectangular coordinate system and a discrimination frame is placed
305
around the platelet cluster. The instrument measures the complexity (right angle scatter), cell size (forward angle scatter), and percentage of platelets demonstrating surface expression of P-selectin and fibrinogen binding. Analytical markers were set in the forward angle scatter channel to divide the negative control sample into two fractions; one containing 98--99% of the platelets, and the other containing the smallest platelets, 1--2% of the total. Those platelets with forward scatter lower than the marker were identified as microparticles. Results were expressed as the percentage of anti-GP Ib antibody-positive platelets, defined as P-selectin- or fibrinogen-positive with a fluorescence intensity exceeding that of 98--99% of control platelets.
Statistical analysis All data are reported as mean F standard deviation. We performed analyses using StatView for Windows J5.0 (SAS institute, North Carolina). The two groups were compared using the nonparametric Mann--Whitney U-test at each time point and differences were considered to be statistically significant when the p-value was less than 0.05.
Results Inhibition of platelet aggregation to ADP by FK633 Inhibition of platelet aggregation was 59 F 10%, 76 F 8.2%, 96 F 3.8%, and 99 F 2.8% at final concentrations of 100 nmol/l, 300 nmol/l, 1 mmol/l, and 3 mmol/l, respectively. Platelet aggregation in PRP was thus inhibited dose-dependently. We therefore decided to administer a dosage of 0.2 mg FK633 to the circuit to give a final concentration of 1 mmol/l.
Blood counts At baseline, no significant differences in differential blood count were observed between the two groups: white blood cells (group F; 4.36 F 0.49 vs. group C; 5.64 F 1.43, p = 0.14), red blood cells (F; 4.63 F 0.1 vs. C; 4.83 F 0.38, p = 0.58), hemoglobin (F; 14.6 F 1.2 vs. C; 15.1 F 1.6, p = 0.67), hematocrit (F; 41.9 F 3.2 vs. C; 43.9 F 4.4, p = 0.49), and platelet counts (F; 193 F 43.9 vs. C; 226 F 52.9, p = 0.31). All values but platelet counts did not change significantly throughout
306
Measures of platelet function at each time point
Variable Platelet count (% of baseline) Platelet aggregation (% of baseline) Beta-thromboglobulin (ng/ml) Thrombin--antithrombin complex (Ag/ml) Bradykinin (pg/ml) Fibrinogen (mg/dl) Fibrinogen binding (% of positive platelets) P-selectin (% of positive platelets) Microparticles (/10000 GP Ib-positive cells)
Group F C F C F C F C F C F C F C F C F C
Baseline 100 100 100 100 464 F 159 317 F 190 2.3 F 1.0 3.7 F 2.0 22.5 F 12.7 32.5 F 2.9 234 F 81 175 F 15 2.7 F 0.4 2.4 F 0.5 0.4 F 0.2 0.4 F 0.3 0.19 F 0.12 0.10 F 0.10
Bag 98.9 F 11.7 90.3 F 12.9 18.7 F 7.4 54.2 F 17.6** 1623 F 1209 1678 F 1358 3.9 F 2.4 2.9 F 0.9 30.2 F 21.8 41.2 F 5.6 158 F 47 122 F 8 3.5 F 1.1 10.0 F 4.8** 30.7 F 7.0 26.3 F 6.8 0.15 F 0.10 0.14 F 0.15
Recirculation time (min) 5
15
30
60
120
93.3 F 10.2 83.4 F 4.6 22.1 F 8.4 71.5 F 25.3** 1663 F 1251 2250 F 1118 3.8 F 1.9 3.3 F 1.2 32.0 F 24.7 40.1 F 9.1 145 F 63 122 F 13 3.4 F 1.6 6.7 F 3.4 28.3 F 3.5 17.0 F 4.9* 0.42 F 0.3 0.13 F 0.1
88.8 F 11.0 76.2 F 4.4 27.5 F 13.1 64.4 F 19.5** 1642 F 1211 2498 F 1049 3.1 F 3.1 4.5 F 2.9 14.1 F 12.3 36.9 F 6.7 148 F 60 121 F 12 2.9 F 0.9 5.5 F 2.0 24.4 F 5.8 17.3 F 5.2* 0.28 F 0.18 0.13 F 0.12
86.8 F 12.4 59.8 F 13.4* 18.0 F 10.7 68.6 F 25.9** 1708 F 1285 2862 F 1134 4.4 F 2.3 5.6 F 4.3 18.0 F 10.8 37.0 F 7.8 143 F 52 125 F 9 2.5 F 1.0 4.6 F 1.9 20.1 F 6.9 19.3 F 3.7 0.30 F 0.26 0.15 F 0.06
83.4 F 13.2 43.6 F 15.8** 21.0 F 11.6 79.8 F 25.7** 1615 F 1134 3952 F 1129* 3.6 F 1.7 7.0 F 4.7 15.6 F 11.5 35.3 F 9.2 154 F 50 124 F 13 2.9 F 0.8 8.1 F 4.5* 16.0 F 5.2 12.8 F 3.4 0.28 F 0.17 0.58 F 0.39
77.4 F 8.7 35.1 F 12.1** 23.5 F 11.1 54.4 F 17.7** 1986 F 1147 5629 F 1711* 6.4 F 1.8 15.6 F 5.7* 15.9 F 8.7 33.5 F 8.5 159 F 56 124 F 7 2.7 F 0.9 11.0 F 4.7** 5.8 F 5.0 15.3 F 5.8* 0.34 F 0.20 3.25 F 3.03
Data are expressed as mean and standard deviation of mean. * p < 0.05 between the control group (C) and the group given FK633 (F) (Mann--Whitney U-test). ** p < 0.01 between the control group (C) and the group given FK633 (F) (Mann--Whitney U-test).
N. Kondo et al. / Thrombosis Research 113 (2004) 303--310
Table 1
N. Kondo et al. / Thrombosis Research 113 (2004) 303--310 recirculation except due to dilution with the circuit contents (data not shown). However, in both groups, significant changes in platelet count oc-
307
curred over time (Table 1; Fig. 1A). In group C, platelet counts after 30-min circulation were significantly lower than those at baseline and signif-
Fig. 1 Effects of FK633 on (A) platelet counts, (B) beta-thromboglobulin, (C) fibrinogen binding, (D) P-selectin expression and (E) microparticle. Data are expressed as triangle and continuous line (group F, recirculated with heparin and FK633) and rhombs and dotted line (group C, recirculated with heparin). Data points represent mean, error bars represent standard deviation. *p < 0.05; **p < 0.01 (comparison using the nonparametric Mann--Whitney U-test at each time points).
308
N. Kondo et al. / Thrombosis Research 113 (2004) 303--310
icantly lower than those in group F at the same sample time.
Platelet aggregation to ADP (Table 1) Platelet aggregation was significantly lower in group F at time point 0 and remained at the same level until the end of recirculation.
Soluble markers (Table 1) In both groups, levels of BTG increased at time point 0. BTG levels in group F remained unchanged thereafter, whereas levels in group C increased with ongoing recirculation (Fig. 1B). TAT levels significantly increased at 120 min after simulated extracorporeal circulation in group C, but no increase was observed in group F. In both groups, levels of fibrinogen decreased at time point 0 and remained unchanged throughout recirculation. In both groups, bradykinin levels increased at time point 0 and decreased with recirculation. However, changes in bradykinin levels did not reach significance.
Flow cytometry In group F, percentage of platelets exhibiting fibrinogen binding remained unchanged throughout recirculation, whereas that in group C increased significantly at time point 0. In group C, platelet fibrinogen binding decreased with recirculation, but increased at 60 and 120 min (Fig. 1C). In both groups, P-selectin expression increased at time point 0. Although P-selectin expression was preserved in group F at 5 min of recirculation, it reduced consistently over time, and fell to levels below those seen in group C by 120 min (Fig. 1D). The percentage of microparticles increased 120 min after simulated extracorporeal circulation in group C (Fig. 1E); however, no increase in microparticles was observed in group F.
Discussion Bleeding complications are a constant problem after cardiac surgery, resulting in 4% of patients requiring reentry into the operating theater after the first 24 h post surgery [1]. Such complications are related to the use of heparin, the onset of fibrinolysis, platelet loss and dysfunction, and a combination of these factors [14]. Platelets are partially activated by heparin and are diluted, destroyed, and modified during CPB because of adhesion to circuit surfaces,
aggregation, and activation [2]. Therefore, after CPB, a reduction in platelet number and function prolongs bleeding times and contributes to postoperative bleeding. Argatroban, a short-acting direct thrombin inhibitor, was used in a patient with heparin induced thrombocytopenia and was found not to have a platelet-activating effect. However, unlike heparin, which can be neutralized with protamine, argatroban has no neutralizing agent [15] and it is therefore impossible to reverse its action at the end of extracorporeal circulation. Therefore, heparin could not currently be replaced by argatroban. In another study, platelets that had become microaggregated due to heparin administration during extracorporeal circulation were preferentially lost from circulation when compared to single platelets [8]. Nonetheless, there is a possibility that, despite activation, platelets can be preserved by temporarily inhibiting their aggregation. Glycoprotein IIb/IIIa inhibitors target the final common pathway of platelet aggregation, which involves the cross-linking of glycoprotein IIb/IIIa complex, a receptor for adhesive plasma proteins, particularly fibrinogen. Previous studies have validated the concept of platelet anesthesia using a GP IIb/IIIa inhibitor to prevent loss of platelet number and function during CPB [9,10,16]. We hypothesized that platelet anesthesia with FK633, a short-acting GP IIb/IIIa inhibitor, functions by keeping activated platelets single. This study revealed that platelet anesthesia with FK633 could preserve platelet numbers to an extent similar to that reported previously [17,18] using other GP IIb/IIIa inhibitors. From the results of BTG levels in the present study, platelets were already activated when blood was drawn from donor to reservoir bag. BTG levels correlated closely with platelet counts, rising during recirculation in group C, while platelet count simultaneously decreased. Difference of BTG levels between the two groups was significant from 60 min, but evident from 5-min recirculation. Mollnes et al. [20] reported that platelet counts and BTG quantification could be used to reliably evaluate the degree of platelet biocompatibility. Although platelet counts are undoubtedly important, as mentioned in the study above, they provide no direct indication of platelet activation. Platelet counts are useful if combined with methods directly reflecting activation [20]. To ascertain the effects of FK633, we performed platelet aggregation and flow cytometry for direct analysis of individual platelets, in a manner similar to that of Mollnes et al. We confirmed our hypothesis based on these results. Results of platelet aggregation to ADP and from flow cytometry of fibrinogen binding indicated that FK633 preserves platelets by inhibition of
N. Kondo et al. / Thrombosis Research 113 (2004) 303--310 fibrinogen binding to GP IIb/IIIa complex. Effects of varying FK633 dose on platelet aggregation were comparable with previous reports [19]. The percentage inhibition in group F actually remained at about 80% although we administered FK633 so as to inhibit submaximally on the basis of the current and previous studies. Musial et al. [18] reported that short disintegrins have a higher affinity for resting platelets than do medium disintegrins. FK633 may also have higher affinity for resting platelets. It is thought that the action of FK633 declined because platelets had already become activated while in the reservoir bag. In group F, fibrinogen did not bind to platelets, whereas in group C fibrinogen binding was seen at time point 0 and decreased rapidly with recirculation. This finding suggested that platelets cross-linked with fibrinogen were consumed from the beginning of recirculation. It appears that the increase in proportion of fibrinogen-positive platelets is based on reduction of platelets due to consumption. P-selectin-positive platelets increased in parallel with BTG levels in both groups at time point 0. These flow cytometry results support the notion that platelets exposed to FK633 exist in the extracorporeal circuit in the activated single state, and that as a result, rapid platelet consumption did not occur at initiation of recirculation. However, platelet consumption did occur over time in group F, suggesting that the activated GP IIb/IIIa complex might cause fibrinogen binding and aggregation during recirculation. In actual cardiac surgery, administration of heparin occurs before initiation of cardiopulmonary bypass. Administration of FK633 before heparin might therefore suppress the activation of platelets by the circuit and by heparin. BTG levels and P-selectin expression indicated that platelet activation during extracorporeal circulation did not occur in group F. FK633 is thought to inhibit an amplification loop in platelet aggregation, involving release of molecules such as ADP and 5hydroxytryptamine (serotonin) from activated platelets, after fibrinogen binding [21]. We had hypothesized that P-selectin-positive platelets would increase during recirculation due to preservation of activated platelets. However, based on these results, the effects of FK633 on platelet preservation result from the inhibition of platelet activation during recirculation. Levels of TAT complex, an indication of plasma thrombin level induced by an activated coagulation system, increased in group C in which only heparin was administered. In group F, when a highly specific antiplatelet agent [19] was added, TAT levels were unchanged, suggesting that thrombin production might also
309
contribute to activation of platelets during simulated ECC. These findings are in accordance with the results obtained by Carroll et al. [22] and by Li et al. [23], which showed inhibition of a-granule release and strong inhibition of thrombin generation with GP IIb/IIIa inhibitors. Bradykinin reflects activation of the contact system. Contact with the artificial surfaces of the extracorporeal circuit activates the intrinsic coagulation pathway in which bradykinin is produced by the kallikrein-kinin system. Differences in bradykinin levels between the two groups did not reach significance, suggesting that the simulated extracorporeal circuit had little influence on the intrinsic coagulation pathway. Plasma levels of fibrinogen tend to decrease as fibrinogen is deposited within the circuit. Since fibrinogen levels remained unchanged despite platelet consumption, the amount of plasma fibrinogen appeared to be more than sufficient. In the present study, the simulated extracorporeal circuit, an entirely in-vitro system, conferred the advantage that results reflected platelet activation due to recirculation alone, because new platelets could not be recruited from the bone marrow, liver, or spleen. In addition, since this system avoided the effects of hepatic or renal metabolism and excretion on soluble markers of platelet activation, levels of markers were considered to reflect the difference between the two groups more clearly. Though further confirmation is needed for the application of this in-vitro study to in-vivo conditions, there are several reports of the perioperative use of short-acting GP IIb/IIIa inhibitors in cardiopulmonary bypass in clinical or animal experiments. Gammie et al. [24] reported that abciximab, when given within 12 h of cardiac surgery, was associated with significantly increased blood loss and transfusion. Abciximab is the Fab fragment of chimeric mouse/human immunoglobulin G antibody. Although its plasma half-life is 26 min, its antiplatelet effect is prolonged and irreversible after administration [25,26]. Bizzarri et al. [1] reported that the infusion of tirofiban hydrochloride immediately before surgical intervention did not adversely affect clinical outcome and had no deleterious effect on postoperative bleeding, maintaining hemoglobin and platelet counts at stable levels. Suzuki et al. [9,10] reported that eptifibatide reduces platelet activation during cardiopulmonary bypass and accelerates restoration of normal bleeding times after operation in a baboon model. The effects of tirofiban and eptifibatide are reversible. On the other hand, FK633 has a reversible antiplatelet effect and a plasma half-life of 0.52 h [19], which
310
N. Kondo et al. / Thrombosis Research 113 (2004) 303--310
is shorter than those of tirofiban and eptifibatide. Therefore FK633 may be able to protect platelets and decrease postoperative bleeding. In conclusion, platelet counts could be maintained by platelet anesthesia with FK633 during extracorporeal circulation. Although FK633 could not completely prevent consumption of platelets, platelets were not activated despite prolonged recirculation. These results suggest that FK633 inhibits the amplification loop by inhibiting the binding of fibrinogen to glycoprotein IIb/IIIa and platelet aggregation.
Acknowledgements The authors thank Ms. Yuko Tsushima for flow cytometry, Mr. Masakatsu Nakagawara (MERA) for creation of perfusion circuits and supervision of recirculation, and Mr. Yuta Inoue and Ms. Maya Ohashi for their excellent technical support. We also are indebted to Dr. Yoshiko Tamai for helpful discussions of the manuscript.
References [1] Bizzarri F, Scolletta S, Tucci E, Lucidi M, Davoli G, Toscano T, et al. Perioperative use of tirofiban hydrochloride (Aggrastat) does not increase surgical bleeding after emergency or urgent coronary artery bypass grafting. J Thorac Cardiovasc Surg 2001;122:1181--5. [2] Woodman RC, Harker LA. Bleeding complication associated with cardiopulmonary bypass. Blood 1990;76:1680--97. [3] Tanaka K, Takao M, Yada I, Yuasa H, Kugasawa M, Deguchi K. Alteration in coagulation and fibrinolysis associated with cardiopulmonary bypass during open heart surgery. J Cardiovasc Anesth 1989;3:181--8. [4] Gu YJ, van Oeveren W, Akkerman C, Boonstra PW, Huyzen RJ, Wildevuur CR. Heparin-coated circuits reduce the inflammatory response to cardiopulmonary bypass. Ann Thorac Surg 1993;55:917--22. [5] Johnell M, Elgue G, Larsson R, Larsson A, Thelin S, Siegbahn A. Coagulation, fibrinolysis, and cell activation in patients and shed mediastinal blood during coronary artery bypass grafting with a new heparin-coated surface. J Thorac Cardiovasc Surg 2002 August;124(2):321--32. [6] De Somer F, Francßois K, van Oeveren W, Poelaert J, De Wolf D, Ebels T, et al. Phosphorylcholine coating of extracorporeal circuits provides natural protection against blood activation by the material surface. Eur J Cardiothorac Surg 2000;18:602--6. [7] De Somer F, Van Belleghem Y, Caes F, Francßois K, Van Overbeke H, Arnout J, et al. Tissue factor as the main activator of the coagulation system during cardiopulmonary bypass. J Thorac Cardiovasc Surg 2002;123:951--8. [8] Muriithi EW, Belcher PR, Rao JN, Chaudhry MA, Nicol D, Wheatley DJ. The effects of heparin and extracorporeal circulation on platelet counts and platelet microaggregation during cardiopulmonary bypass. J Thorac Cardiovasc Surg 2000;120:538--43.
[9] Suzuki Y, Hillyer P, Miyamoto S, Niewiarowski S, Sun L, Rao AK, et al. Integrilin prevents prolonged bleeding times after cardiopulmonary bypass. Ann Thorac Surg 1998;66: 373--81. [10] Suzuki Y, Malekan R, WilliamHanson CW, Niewiarowski S, Sun L, Rao AK, et al. Platelet anesthesia with nitric oxide with or without eptifibatide during cardiopulmonary bypass in baboons. J Thorac Cardiovasc Surg 1999;117:987--93. [11] Ruoslahti E. Integrins. J Clin Invest 1991;87:1--5. [12] Yamazaki M, Uchiyama S, Iwata M. Measurement of platelet fibrinogen binding and P-selectin expression by flow cytometry in patients with cerebral infarction. Thromb Res 2001;104:197--205. [13] Michelson AD. Flow cytometry: a clinical test of platelet function. Blood 1996;87:4925--36. [14] Khuri SF, Valeri CR, Loscalzo J, Weinstein MJ, Birjiniuk V, Healey NA, et al. Heparin causes platelet dysfunction and induces fibrinolysis before cardiopulmonary bypass. Ann Thorac Surg 1995;60:1008--14. [15] Levy JH. Pharmacologic preservation of the hemostatic system during cardiac surgery. Ann Thorac Surg 2001;72: S1814--20. [16] Hiramatsu Y, Gikakis N, Anderson III HL, Gorman III JH, Marcinkiewicz C, Gould RJ, et al. Tirofiban provides ‘‘platelet anesthesia’’ during cardiopulmonary bypass in baboons. J Thorac Cardiovasc Surg 1997;113:182--93. [17] Bernabei A, Gikakis N, Kowalska MA, Niewiarowski S, Edmunds Jr LH. Iloprost and echistatin protect platelets during simulated extracorporeal circulation. Ann Thorac Surg 1995;59:149--53. [18] Musial J, Niewiarowski S, Rucinski B, Stewart GJ, Cook JJ, Williams JA, et al. Inhibition of platelet adhesion to surfaces of extracorporeal circuits by disintegrins. RGD-containing peptides from viper venoms. Circulation 1990;82(1): 261--73. [19] Aoki T, Cox D, Senzaki K, Seki J, Tanaka A, Takasugi H, et al. The anti-platelet and anti-thrombotic effects of FK633, a peptide-mimetic GPIIB/IIIA antagonist. Thromb Res 1996;81:439--50. [20] Mollnes TE, Videm V, Christiansen D, Bergseth G, Riesenfeld J, Hovig T. Platelet compatibility of an artificial surface modified with functionally active heparin. Thromb Haemost 1999;82:1132--6. [21] Hong J, Larsson A, Ekdahl KN, Elgue G, Larsson R, Nilsson B. Contact between a polymer and whole blood: sequence of events leading to thrombin generation. J Lab Clin Med 2001;138:139--45. [22] Carroll RC, Wang XF, Lanza F, Steiner B, Kouns WC. Blocking platelet aggregation inhibits thromboxane A2 formation by low dose agonists but does not inhibit phosphorylation and activation of cytosolic phospholipase A2. Thromb Res 1997;88:109--25. [23] Li N, Wallen NH, Savi P, Herault JP, Herbert JM. Effects of a new platelet glycoprotein IIb/IIIa antagonist, SR121566, on platelet activation, platelet--leukocyte interaction and thrombin generation. Blood Coagul Fibrinolysis 1998;9: 507--15. [24] Gammie JS, Zenati M, Kormos RL, Hattler BG, Wei LM, Pellegrini RV, et al. Abciximab and the risk of bleeding during emergency cardiac operations. Ann Thorac Surg 1999; 67:292--3. [25] Coller BS. GPIIb/IIIa antagonists: pathophysiologic and therapeutic insights from studies of c7E3 Fab. Thromb Haemost 1997;78:730--5. [26] Kleiman NS. Pharmacokinetics and pharmacodynamics of glycoprotein IIb--IIIa inhibitors. Am Heart J 1999;138: 263--75.