Platelet Loss during Experimental Cardiopulmonary Bypass and Its Prevention with Prostacyclin

Transcription

1 Platelet Loss during Experimental Cardiopulmonary Bypass and Its Prevention with Prostacyclin John R. Plachetka, Pharm.D., Neal W. Salomon, M.D., Douglas F. Larson, M.S., and Jack G. Copeland, M.D. ABSTRACT Prostacyclin (PGI,), a newly discovered short-acting prostaglandin that inhibits platelet aggregation, was evaluated as an agent for prevention of cardiopulmonary bypass-induced thrombocytopenia. Ten adult, splenectomized greyhounds were divided into three treatment groups prior to beginning 120 minutes of partial cardiopulmonary bypass. Group 1 animals received 300 units of heparin per kilogram of body weight, Group 2 animals received 300 units of heparin per kilogram plus PGI,, 1.5 pg per minute, and Group 3 animals received 300 units of heparin per kilogram plus PGI, 3.0 pg per minute. Bypass and PGI, infusion were started simultaneously. Mean platelet counts of each group at 5 minutes were approximately 40% of prebypass levels. Additional platelet loss was seen in Groups 1 and 2 at 30, 60, and 120 minutes. However in Group 3, platelet counts at 30 and 60 minutes were essentially unchanged from prebypass levels. At 30, 60, and 120 minutes of cardiopulmonary bypass, the differences between Groups 1 and 3, and 2 and 3 are highly significant (p < 0.01). We conclude that PGI, is an effective agent for preserving platelet levels during experimental cardiopulmonary bypass. Furthermore, it is possible that platelet loss during cardiopulmonary bypass may be caused, in part, by an imbalance between PGI, and thromboxane A,, which results in excessive platelet adhesion and aggregation. Thrombocytopenia is a common problem after cardiopulmonary bypass operation [5, 15, 21, From the Division of Cardiovascular and Thoracic Surgery, and the Department of Pharmacy Practice, University of Arizona Health Sciences Center, Tucson, AZ. Supported by a University of Arizona Biomedical Research Support Grant. Accepted for publication Nov 14, Address reprint requests to Dr. Plachetka, Department of Pharmacy Practice, College of Pharmacy, University of Arizona, Tucson, AZ and predisposes patients to potentially serious hemorrhagic events. In an effort to prevent this occurrence, investigators have turned to pharmacological platelet inhibitors like aspirin, dipyridamole, and prostaglandin E, (PGE,) with varying degrees of success [2, 17J. While aspirin (65 mg per kilogram of body weight) has been shown to be ineffective in preventing platelet loss, dipyridamole (4 mg per kilogram) moderately reduces platelet losses in dogs [17]. PGE,, a prostaglandin with antiaggregatory activity, has no consistently beneficial effects on platelet counts during cardiopulmonary bypass in monkeys [2], but can reduce platelet loss during ex vivo recirculation of human and monkey blood through a pump oxygenator apparatus [l]. Therefore, the need still exists for an agent able to prevent platelet loss during cardiopulmonary bypass. Finding such an agent also should aid in the elucidation of the mechanism whereby platelets are lost during the procedure. Prostacyclin (PG12) is a potentially useful agent for platelet preservation during cardiopulmonary bypass. While it is the most potent antiaggregatory agent yet discovered, the antiplatelet action of PGI, is short-lived [191. Thus, normal platelet function returns soon after administration of the agent is discontinued [14]. Woods and colleagues [241 demonstrated that PGI, alone, without the use of heparin, prevents platelet loss during hernodialysis in dogs. We report here our experience using PGI, during cardiopulmonary bypass in dogs and propose a theory on the relationship between this agent and platelet loss during this procedure. Material and Methods Ten adult greyhound dogs weighing 23 to 28 kg were anesthetized with pentobarbital, 30 mg per kilogram of body weight. The dogs were by John R. Plachetka

2 59 Plachetka et al: Platelet Loss during Experimental CPB ventilated with a volume ventilator (Bird model Mark VII) through a cuffed endotracheal tube throughout the experiment. A splenectomy was performed on all dogs immediately prior to cardiopulmonary bypass to prevent a flux of blood components between the spleen and the intravascular space [3]. The dogs were systemically heparinized with beef lung heparin, 300 units per kilogram, from the Upjohn Company. Cannulation of the external jugular vein and bilateral femoral veins provided venous return, and arterial perfusion was accomplished through a femoral artery. All dogs were perfused at a blood flow rate of 70 ml per kilogram with a barely occlusive, calibrated double-roller Travenol pump (model 500) for 120 minutes. Oxygen, carbon dioxide, and heat exchange were provided by a Harvey oxygenator (model H-100). The oxygen to blood flow ratio was 2 : 1, and the perfusate temperature was maintained at 37 C. The oxygenator and tubing were primed with 1,500 ml of lactated Ringer s solution. A Bentley blood filter (model PF-427) was placed in the arterial line for emboli filtration. Solutions of PGI, were freshly prepared using PGI, sodium salt* (1.14 mg = 1.0 mg free acid *Kindly supplied by Dr. John Pike, The Upjohn Company, Kalamazoo, MI. Fig 1. Platelet counts, expressed as percent of prebypass values, of Groups 1, 2, and 3 plotted versus time. Differences between Groups 1 and 3 and between Groups 2 and 3 are significant at 30, 60, and 120 minutes. Values are mean f 1 standard deviation. (* = p < 0.001; A = p < 0.01.) equivalents PGI,) and BuffAR ph 10 (Mallinckrodt). The final concentration of the solution was 10 pg of PGI, per milliliter. The solution was infused with an IVAC constant infusion pump (model 200) into the venous port of the oxygenator. Group 1 animals (n = 4) did not receive any PGI,. Group 2 (n = 3) animals received 1.5 pg of PGI, per minute. Group 3 animals (n = 3) received 3.0 pg of PGI, per minute. Infusion of PGI, was begun at the time bypass was instituted. Data were collected immediately before bypass and 5, 30, 60, and 120 minutes into bypass. Microhematocrits were determined with an Internation Micro-Capillary Centrifuge model MB. Platelet counts (phase contrast microscopy) were performed by the clinical laboratory without knowledge of the experimental procedure [4]. Radial artery pressures were monitored with a Trantec pressure transducer (model 800) connected to an Electronics for Medicine monitor (model VR-6). A two-tailed, unpaired Student s t test was applied to the data to determine statistical significance. Results Results at 5 Minutes Platelet counts, expressed as percent of prebypass levels (mean k standard deviation), initially fell in all three groups (Fig 1). Although platelet counts could be expected to fall with the clear fluid prime, the observed decline exceeded that expected from dilution alone (Fig Pre-Bypass 5 30 Time (minutes) 120

3 60 The Annals of Thoracic Surgery Vol30 No 1 July 1980 in either Group 1, 13 f 9%, or Group 2, 27 k 12% ( p < 0.01). However, it did fall below the previously mentioned dilution range. Bypass Time (minutes) Fig 2. Platelet counts, expressed as percent of prebypass values, of Groups 1 (0) and 3 (X) plotted versus time. Dotted area shows the expected effect of dilution alone on platelet counts. Differences between Groups 1 and 3 at 30 minutes and at 60 and 120 minutes are significant (p < and p < 0.01, respectively). Values are mean k 1 standard deviation. 2). The dilution factor was calculated from the mean changes in hematocrits at 5 minutes from prebypass levels. There were no significant differences between groups with respect to 5-minute platelet counts. Results at 30 Minutes The platelet counts in Group 1 continued to fall to a mean value of 20 f 9% of the prebypass level. No significant change occurred between 5 minutes and 30 minutes in Group 2. However, compared with control, mean platelet counts in Group 3 rose significantly during this interval to a mean value of 68 f 4% of the prebypass level, within the range of platelet counts expected from dilution alone. These values in Group 3, unlike Group 2, were significantly higher than in Group 1 ( p < 0.001). Results at 60 Minutes While the platelet counts of Groups 1 and 2 remained at approximately 20% of control, the mean value of Group 3, 66 f 12% of control, remained within the range expected by dilution alone. The differences between Groups 1 and 3 and between Groups 2 and 3 were significant (p < 0.01). Results at 120 Minutes The mean platelet counts relative to control for Group 3, 47 f 8%, remained higher than those Comment At normal concentrations, PGI, inhibits platelet aggregation and at higher concentrations, platelet adhesion [14, 18, 19, 201. Furthermore, even higher levels of PGI, can cause deaggregation of recently formed platelet clumps [lo, 121. These effects of PGI, seem to be mediated through the stimulation of adenylate cyclase, which, in turn, leads to increased levels of platelet camp [7-9, 231. Prostacyclin biosynthesis occurs mainly in healthy pulmonary and vascular endothelial cells [4, 11, 231 where PGIz synthetase converts endogenous precursors or platelet-liberated endoperoxides to PGIz [191. Since its blood half-life is only 2 to 3 minutes [6, 131, these tissues must produce PGIz continuously to maintain effective, circulating concentrations [4]. In contrast to PG12, prostaglandin endoperoxides and principally thromboxane A, (TXA,) have an anti-camp activity in platelets (Fig 3) [7-9, 231. These substances, which are produced by platelets and damaged blood vessels, promote platelet adhesion and aggregation by decreasing platelet camp [7-9, 19, 231. Certainly, collagen, epinephrine, adenosine diphosphate, and other substances also contribute to platelet homeostasis, but recent evidence indicates that PGI, and TXAz play key roles in controlling platelet function in vivo [7-9, 231. With respect to the involvement of PGI, and TXA2 in platelet homeostasis, it is presently thought that under normal conditions, circulating levels of PGIp are sufficient to offset the pro-aggregating influence of TXA, and prevent spontaneous aggregation and adherence of platelets to endothelium. However, when the endothelium is damaged, platelets congregate at the site and undergo a shape change, releasing the contents of their storage granules. Concentrations of endoperoxides and TXA2 increase as the thrombus grows, further promoting platelet aggregation and adhesion. In general, this process has only a local effect, due to the short biological half-life of TXA,, the circulating levels of PGI,, and the conversion of platelet-

4 ,*\ 61 Plachetka et al: Platelet Loss during Experimental CPB Arachidonic Acid t /, (Prostacyclin Synthetase) (lhromboxane Synthetase) PGI, r/ \ / 3 Excess TXA, 1) t CAMP 1) &CAMP 2) Anti-Aggregatory 2) Pro-Aggregatory 3) Major Sources-Healthy 3) Major Sources-Platelets, Pulmonary and Vascular Damaged Blood Vessels Endothelial Cells 4) Half-life-30 sec 4) Half-life-2-3 minutes Deaggregation Normal Platelet Homeostasis Aggregation Fig 3. Although prostacyclin (PGI,) and thromboxane A, (TXAJ are biosynthesized from common precursors, they are produced by different tissues and have opposing effects on platelets. It has been proposed that a balance system consisting of TXAP and PGI,, as shown here, plays a major role in controlling platelet activity. liberated endoperoxides to PGI, by normal vascular endothelium surrounding the lesion. It is a system capable of quickly establishing hemostasis without undue obstruction of blood flow by limiting the size of the thrombus. During extracorporeal circulation, these basic mechanisms are altered (Fig 4). Since PGI, is provided primarily by the lung [ll, 131 and healthy vascular endothelium, blood is shunted around the major site of biosynthesis. The effect of any residual production of PG12 is diluted by the increased plasma volume required for the bypass procedure. Also, platelet contact with artificial surfaces [5, 151 and the presence Fig 4. Effects of cardiopulmonary bypass on the balance between prostacyclin (PGI,) and thromboxane A2 (TXA,). Increased levels of TXA, and decreased levels of PGI, promote excess platelet adhesion and aggregation and result in thrombocytopenia. Administration of exogenous PG12 during cardiopulmonary bypass reestablishes the balance and prevents platelet loss. of a blood-gas interface [17] have been proposed as factors precipitating platelet release of pro-aggregating endoperoxides and TXA,. We propose that there is an imbalance between PGI, and TXA, in the extracorporeal system and that it favors platelet adhesion and aggregation. This imbalance could be caused by decreasing levels of PG12, increased production of TXA,, or, most likely, a combination of the two. The results of our experiment support this theory. Because we began infusions of PGIz into the venous port of the oxygenator precisely when bypass was instituted, we would expect to see little, if any, effect of the PGI, on 5- minute platelet counts. As shown in Figure 1, there were no significant differences among the three groups at that time. While animals in the control group (Group 1) had a further decrease in platelet numbers at 30 minutes to 20% of prebypass levels, animals receiving 3 pg of PG12 per minute (Group 3) had platelet counts in the range expected by dilution alone. In other words, platelet counts of these animals were indistinguishable from prebypass values corrected for dilution. This difference is highly significant ( p < 0.001) and deserves further explanation. Prostacyclin, while being a potent inhibitor of platelet aggregation, also possesses the TTXA, 1) Pump Hardware 8 Blood-Gas Interface Activates Platelets 2) Numbers of Platelets Stimulated 1) Short Half-life 2) Exclusion of PGI, From Lungs 31 Dilution 4j Extracorporeal System Unable to Synthesize PGI, Deaggregation Normal Platelet Homeostasis Excess Aggregation

5 62 The Annals of Thoracic Surgery Vol30 No 1 July 1980 ability to disperse platelet clumps. Gryglewski and colleagues [lo] demonstrated that PGI, disintegrates platelet aggregates, even after three hours. As the concentration of PGI, in our system increased, it is probable that platelet clumps formed during the period when the concentration of PGI, was inadequate deaggregated into individual platelets. Platelet counts would seem to rise, as shown in our data, but there is no real change in numbers of platelets, only in their location. Although PGI, was not yet identified at the time that Mielke and coworkers [17] did their study, their animals given the phosphodiesterase inhibitor, dipyridamole, showed a similar, albeit, less dramatic effect. The lack of effect of PGI, on Group 2 animals was unexpected, but is explained readily. In our original design, we chose that dose of PGIz because it had been used successfully to prevent thrombocytopenia during hemodialysis [24]. However, this did not take into account the important dilution that occurs during cardiopulmonary bypass. Thus, the actual circulating concentration was only 60 to 70% of expected. Doubling the dose of PGI, produced dramatic platelet preservation, as evidenced by Group 3 values. Longmore and associates [16] reported in 1979 that PGI,, alone or in combination with heparin, prevented thrombocytopenia in nonsplenectomized greyhounds undergoing two hours of complete cardiopulmonary bypass. Platelet function returned to normal within 30 minutes of discontinuing the infusions. The dose of PGI, was five to ten times the dose used in our study. Prostacyclin is effective in maintaining platelet counts during experimental cardiopulmonary bypass. The hypothesized relationship between alterations in the PGI, : TXA, ratio and thrombocytopenia seen after this procedure has already been expressed. Since PGI, is still classified as an investigational drug, the results of future clinical trials will help to define its role in the prevention of platelet loss during human cardiopulmonary bypass. References 1. Addonizio VP, Macarak EJ, Niewiarowski S, et al: Preservation of human platelets with pros- taglandin El during in vitro simulation of cardiopulmonary bypass. Circ Res 44:350, Addonizio VP, Strauss JF, Macarak EJ, et al: Preservation of platelet number and function with prostaglandin E, during total cardiopulmonary bypass in rhesus monkeys. Surgery 83:619, 3. Aster R: Pooling of platelets in the spleen: role in the pathogenesis of "hypersplenic" thrombocytopenia. J Clin Invest 45:645, Brecher G, Cronkite E: Morphology and enumeration of human blood platelets. J Appl Physiol 3:365, DeLeval MR, Hill JD, Mielke CH, et al: Blood platelets and extracorporeal circulation. J Thorac Cardiovasc Surg 69:144, Dusting GJ, Moncada S, Vane JR: Disappearance of prostacyclin (PGI,) in the circulation of the dog (proceedings). Br J Pharmacol 62:414P, 7. Gorman RR: Modulation of human platelet function by prostacyclin and thromboxane AZ. Fed Proc 38:83, Gorman RR, Bunting S, Miller OV: Modulation of human platelet adenylate cyclase by prostacyclin (PGX). Prostaglandins 13:377, Gorman RR, Fitzpatrick FA, Miller OV: Reciprocal regulation of human platelet CAMP levels of thromboxane A, and prostacyclin, in Advances in Cyclin Nucleotide Research. Edited by WJ George, LJ Ignahro. New York, Raven Press,, VO~ 9, pp Gryglewski RJ, Korbut R, Ocetkiewicz A: Deaggregatory action of prostacyclin in vivo and its enhancement by theophylline. Prostaglandins 15:637, 11. Gryglewski RJ, Korbut R, Ocetkiewicz A: Generation of prostacyclin by lungs in vivo and its release into arterial circulation. Nature 273:764, 12. Gryglewski RJ, Korbut R, Ocetkiewicz A: Reversal of platelet aggregation by prostacyclin. Pharmacol Res Commun 10:185, 13. Gryglewski RJ, Korbut R, Ocetkiewicz A, et al: Lungs as a generator of prostacyclin-hypothesis on physiologic significance. Naunyn-Schmiedebergs Arch Pharmacol304:45, 14. Gryglewski RJ, Szczeklik A, Nizankowski R: Anti-platelet action of intravenous infusion of prostacyclin in man. Thromb Res 13:153, 15. Kalter RD, Saul CM, Wetstein L, et al: Cardiopulmonary bypass: associated hemostatic abnormalities. J Thorac Cardiovasc Surg 77:427, Longmore DB, Gueirrara D, Bennett G, et al: Prostacyclin: a solution to some problems of extracorporeal circulation. Lancet 1:1002, Mielke CH, DeLeval MR, Hill JD, et al: Drug

6 63 Plachetka et al: Platelet Loss during Experimental CPB influence on platelet loss during extracorporeal circulation. J Thorac Cardiovasc Surg 66:845, Moncada S, Korbut R, Bunting S, et al: Prostacyclin is a circulating hormone. Nature 273:767, 19. Moncada S, Vane JR: Unstable metabolites of arachidonic acid and their role in hemostasis and thrombosis. Br Med Bull 34:129, 20. Moncada S, Vane JR: The role of prostacyclin in vascular tissue. Fed Proc 38:66, Pike OM, Marquiss JE, Weiner RS, et al: A study of platelet counts during cardiopulmonary bypass. Transfusion 12:119, Salzman WE: Blood platelets and extracorporeal circulation. Transfusion 3:274, Tateson JE, Moncada S, Vane JR: Effects of prostacyclin (PGX) on cyclic AMP concentrations in human platelets. Prostaglandins 13:839, Woods HF, Ash G, Western MJ, et al: Prostacyclin can replace heparin in hemodialysis in dogs. Lancet 2:1075, Notice from the Southern Thoracic Surgical Association The Twenty-seventh Annual Meeting of the Southern Thoracic Surgical Association will be held at the Greenbriar, White Sulphur Springs, WV, Nov 13-15,1980. There will be a $100 reg- N Redington Beach, FL istration fee for nonmember physicians except Applications for membership should be completed by Sept 1, 1980, and forwarded to E. P. Smith, Jr., M.D., 425 S Bath Club Blvd, for guest speakers, authors and co-authors on Richard B. McElvein, M.D. the program, and residents. Secretary- Treasurer