| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
J Thorac Cardiovasc Surg 1996;112:523-530
© 1996 Mosby, Inc.
CARDIOPULMONARY BYPASS, |
Supported by Bayer AG, Leverkusen, Germany.
Received for publication April 6, 1995 Accepted for publication Sept. 8, 1995. Address for reprints: C. R. H. Wildevuur, MD, PhD, Department of Thoracic Surgery, Onze Lieve Vrouwe Gasthuis, P.O. Box 95500, 1090 HM Amsterdam, The Netherlands.
Abstract
Prophylactic aprotinin therapy has become a popular method to reduce bleeding associated with cardiac operations. Today essentially two dose regimens are used, a high-dose regimen with administration throughout the complete operative procedure and a low-dose regimen with administration only during bypass. In unblinded studies both regimens were found to be equally effective. This double-blind placebo-controlled study in 115 patients undergoing elective coronary artery bypass grafting was done to confirm these results without potential investigator bias. Intraoperative hemoglobin loss was significantly reduced (p< 0.01) by 42% in the high-dose group and by 17% in the low-dose group compared with loss in control subjects. Blood loss 6 hours after operation was 377 ml in the low-dose and 266 ml in the high-dose group compared with 630 ml in the placebo group (p< 0.05 and p< 0.001, respectively). The average number of transfusions with packed red blood cells was reduced 31% in the low-dose group and 45% in the high-dose group, but the reductions were not significant. In a subgroup of patients, markers for coagulation and fibrinolysis were studied to investigate whether a different extent of activation existed. Fibrinolysis as measured by D-dimer levels was completely inhibited by the high-dose regimen, but was only partly suppressed in the low-dose group as compared with findings in the placebo group. Thrombin generation during cardiopulmonary bypass as reflected by F1+2levels was lower in patients treated with aprotinin, but the difference was not significant. Concentrations of thrombin inactivated by antithrombin III were not different between the groups. The observation that low-dose aprotinin significantly improved hemostasis but did not inhibit hyperfibrinolysis supports our previous finding that low-dose aprotinin mainly protects platelet adhesive function. The better result obtained with high-dose aprotinin may indicate the contribution of hyperfibrinolysis to bleeding after cardiopulmonary bypass. Because high-dose aprotinin is administered outside the period of full heparinization and might therefore increase the risk of thromboembolic complications, we propose a modification of the low-dose schedule to increase aprotinin levels sufficient for plasmin inhibition before release of the aortic crossclamp. (J THORAC CARDIOVASC SURG1996;112:523-30)
Postoperative bleeding is a common problem in cardiac operations and cardiac surgeons have become increasingly aware of the risks associated with transfusion of homologous blood products. This has prompted a search for methods to improve hemostasis in cardiac operations. Of several pharmacologic interventions, perioperative treatment with the antiprotease aprotinin has gained popularity in Europe. Initial experience with aprotinin was obtained with the 6 million kallikrein inhibiting units (KIU), or high-dose, schedule.
1 Numerous reports have confirmed its efficacy and safety. 2-4
There remain, however, concerns about possible thromboembolic side effects and early graft thrombosis with administration of aprotinin outside the period of full heparinization. 5-7 To avoid exposure of the patient to aprotinin during unheparinized periods, Wildevuur and associates 8 introduced the 2 million KIU, or low-dose, regimen in which aprotinin is added solely to the pump prime. Results from nonblinded studies indicate that the efficacy of both regimens is equal. 9-11 This finding has been related to the inhibition of the initial effect of cardiopulmonary bypass (CPB) on platelet hemostatic function, which was shown to be the same for both regimens. 9 The present double-blind placebo-controlled study was designed to demonstrate the comparable effectiveness of the low- and high-dose schedules in a large group of patients.
In a subgroup of patients we measured F1+2 and thrombin–antithrombin III (TAT) complex levels as markers for thrombin generation and D-dimer levels as a marker of fibrinolysis to establish whether the two aprotinin regimens had different effects on these components of the hemostatic balance.
Patients and methods
After informed consent was obtained 115 patients scheduled for elective coronary artery bypass grafting were entered into the study. The study protocol was approved by the scientific and ethical committees of our hospital.
The following exclusion criteria for entry to the study were used: history of previous cardiac operation, possible exposure to aprotinin in the past, allergy or clotting disorder, severe cardiac failure (ejection fraction <25%), and impaired renal function (serum creatinine level >200 µmol/L). A large proportion of the patients were treated with 80 mg acetylsalicylic acid or 100 mg carbasalate calcium daily. These drugs were stopped on the day of admission to the hospital.
Study drug administration
The blinded study medication was supplied by Bayer AG (Leverkusen, Germany) in boxes containing 12 bottles with 50 ml solution. Eight bottles were marked "infus" and four were marked "pump." In the placebo group all bottles contained saline solution; in the low-dose group only the bottles marked "pump" contained 500,000 KIU aprotinin and the remaining bottles contained placebo; and in the high-dose group all bottles contained 500,000 KIU aprotinin.
The solution from the "infus" bottles was transferred into a transfusion bag that was connected to an Ivac infusion pump (Ivac Corporation, San Diego, Calif.). The infusion was started when the first skin incision was made at a rate of 400 ml/hr during 30 minutes to deliver the loading dose of 2 million KIU. During the remainder of the operation the infusion rate was set at 50 ml/hr (500,000 KIU aprotinin/hour). The solution from the "pump" bottles was transferred to the priming volume of the extracorporeal circuit. The randomization code was kept by Bayer AG. The code was broken after data acquisition was complete and verified.
Anesthesia and CPB
Anesthesia comprised premedication with lorazepam; induction with sufentanyl, pancuronium bromide, and etomidate or propofol; and maintenance with a continuous infusion of sufentanyl. An infusion of dopamine and nitroglycerin was routinely started at the termination of CPB.
The extracorporeal circuit contained either a William-Harvey HF 5400 (Bard, Tustin, Calif.) or an Avecor Ultrox I (Avecor, Plymouth, Minn.) oxygenator primed with Ringer's lactate 2000 ml, 20% human albumin 200 ml, 20% mannitol 100 ml, 8% sodium bicarbonate 50 ml, heparin 50 mg, and cefamandole 2 gm. Before cannulation, heparin (3 mg/kg) was given. Predonation was done whenever possible. During bypass an additional 50 mg of heparin was given every hour irrespective of the activated clotting time (ACT). Heparin was neutralized with protamine sulfate in a 1:1 ratio with the initial dose. Additional dosage of protamine was guided by results of repeat ACT tests after the residual volume in the extracorporeal circuit was returned to the patient. Flow ranges were from 2.0 to 2.4 L/m2 per minute during moderate hypothermia (28º to 32º C). Cardioplegia was achieved with ice-cold crystalloid cardioplegic solution infused in the ascending aorta after clamping.
Blood loss
To calculate the intraoperative hemoglobin loss all gauzes were weighed, washed with 1 L saline solution, and the fluid content sampled for hemoglobin concentration. The volume of wasted suction fluid was measured and, after vigorous shaking, sampled for measurement of hemoglobin concentration. With these data we calculated the intraoperative hemoglobin loss. The volume of mediastinally shed blood was measured 6 and 24 hours after operation.
Transfusion of blood products
Packed red blood cells were given during CPB when the hematocrit value dropped below 0.20. After CPB and in the intensive care unit (ICU) packed cells were given when the hematocrit value decreased to less than 0.25. Fresh frozen plasma (FFP) was only given to correct clotting disorders; that is, in the operating room FFP was given when there was insufficient clotting despite an ACT value that had returned to the normal range. FFP was given in the ICU when there was excessive bleeding and normal activated partial thromboplastin time values were present.
Blood sampling
In 14 patients in each group markers for fibrinolysis and coagulation were studied. Blood samples were drawn from a central venous line or the extracorporeal circuit after induction of anesthesia; 5 minutes after onset of extracorporeal circulation; and 10 minutes before and 5 minutes after release of the aortic crossclamp. The samples were collected in tubes containing sodium citrate (final concentration 1.05 mmol/L), centrifuged at 3000 rpm for 10 minutes, and the obtained plasma stored in aliquots at -20º C.
By enzyme-linked immunosorbent assay techniques the concentrations of TAT complex, prothrombin fragment F1+2 (Behringwerke, Marburg, Germany), and D-dimer (Boehringer-Mannheim, Stago, Desnieres, France) were measured. The ratio between the initial hematocrit value and the hematocrit value at the sample point was used to correct the concentrations for hemodilution.
Aprotinin concentrations were measured (Dr. Lemm, Bayer AG, Wuppertal, Germany) in samples drawn 5 minutes after the start of CPB and after administration of protamine according to the method of Müller-Esterl and colleagues. 12
Statistical analysis
Because data of the intraoperative and postoperative blood loss were not normally distributed a log transformation was done. The differences between the groups for these and other continuous data were tested with analysis of variance. The Bonferroni-Holm method was used to correct for the multiple comparisons artifact. The data of F1+2, TAT, and D-dimer measurements were tested with analysis of variance for repeated measurements. Categorical data were tested for statistical significance with the 
2 test. Correlations were calculated with the Spearman rank correlation test. A p value less than 0.05 was considered significant. All analyses were done with SPSS software (SPSS Inc., Chicago, Ill.).
Results
Patients
One hundred fifteen patients were randomized. Three patients were excluded: one patient from the placebo group in whom concurrent with the coronary artery bypass grafting a small left ventricular aneurysm was resected, another patient from the placebo group who had excessive postoperative bleeding because of a broken suture, and one patient from the high-dose group who had dense pericardial adhesions resembling those found in a reoperation.
The demographic and operative data of the three groups are shown in Table I. The groups did not differ except for age, which was significantly lower in the placebo group. However, age did not show a correlation with postoperative blood loss (p > 0.1 in each group, Spearman rank correlation), and the difference was therefore considered not to be of importance.
|
Table II). The volume of mediastinal drainage was reduced significantly in the low- and high-dose groups compared with that in the placebo group at 6 and 24 hours after operation (p < 0.02 and p < 0.05 in the low-dose group and p < 0.001 and p < 0.001 in the high-dose group, respectively). There were no significant differences in intraoperative and postoperative blood loss between the low-and high-dose groups.
|
Blood product use
Approximately three quarters of the patients received blood products (Table III) The reductions in the number of packed cells given after the operation in the low- and high-dose groups were almost significant (p = 0.07 and 0.08, respectively). Also, the reductions in FFP transfusion in the ICU were almost significant (p = 0.05 for the low-dose group and p = 0.09 for the high-dose group). The total amount of FFP given perioperatively was significantly reduced by treatment with 2 million units aprotinin (p < 0.05).
|
|
|
|
Clinical performance
No significant differences were found in duration of artificial ventilation (mean 20 hours) or ICU stay (mean 1.2 days). Perioperative diuresis and fluid balances did not differ significantly between the groups. Creatinine levels were slightly increased in all groups at the fourth postoperative day compared with the preoperative values, but not significantly. Renal failure was not observed in any patient. Perioperative myocardial infarction was seen in five patients in the placebo group, one patient in the low-dose group, and four patients in the high dose group (p = 0.40).
In four patients (4%; three from the placebo group and one from the low-dose group), a rethoracotomy was done because of bleeding. In three patients, one from the placebo group and two from the high-dose group, a rethoracotomy was done because of myocardial ischemia. Two of these patients received an additional venous graft to a vessel that was already grafted with a thoracic artery. At inspection the left internal thoracic artery anastomoses were patent and there was free flow from these grafts, precluding graft thrombosis as the cause for the ischemia. Two patients, one from the placebo and one from the high-dose group, required postoperative support with an intra-aortic balloon pump.
No other major complications or mortality occurred.
Discussion
In this large double-blind trial of aprotinin in elective coronary artery bypass grafting both low-dose (2 million KIU) and high-dose (6 million KIU) aprotinin improved hemostasis, but less intraoperative bleeding was seen with the high-dose schedule. In our previous studies we showed that the bulk of hemoglobin loss occurs during the operative procedure and that postoperative hemoglobin loss accounts for just one quarter to one third of the total hemoglobin loss. 13 Therefore the better efficacy of high-dose aprotinin in this series is reflected by the greater reductions in intraoperative hemoglobin loss, that is, 17% in the low-dose group and 42% in the high-dose group.
The high-dose schedule was initially engineered to obtain a continuous high plasma concentration of aprotinin that would inhibit complement activation through kallikrein formation during CPB and consequently reduce the whole body inflammatory reaction. 1 Because the results of this first study indicated that aprotinin had no effect on complement activation but substantially improved hemostasis, various mechanisms have been proposed to explain the hemostatic efficacy of aprotinin in cardiac operations: inhibition of hyperfibrinolysis, 14,15 inhibition of intrinsic coagulation, 16 preservation of platelet adhesive capacity, 1,17,18 modulation of endothelial eicosanoid synthesis, 19 and protection of platelets from inhibition by heparin. 20 The rationale to determine the mechanism involved is that the administration of aprotinin could then be rationally tailored to obtain an optimal efficacy with a low risk of side effects.
Considering hyperfibrinolysis as the main factor of impaired hemostasis, Royston 21 advocated increasing the initially recommended dose of aprotinin. Wildevuur and associates, 8 on the other hand, recommended tailoring the dose of aprotinin to preserve primarily the platelet adhesive function by giving only a 2 million KIU pump prime dose. In a subpopulation of our study group we examined the effect of low- and high-dose aprotinin on the activity of the coagulation and fibrinolytic systems. Coagulation activity as reflected by F1+2 and TAT complexes increased significantly during CPB, with marked increases after the start of CPB and after release of the aortic crossclamp. F1+2 levels were lower during operation in the aprotinin-treated groups, but the difference did not reach statistical significance. Fibrinolytic activity, as reflected by D-dimer levels, which increased slowly during bypass and increased sharply after release of the aortic crossclamp, correlated inversely with the plasma concentrations of aprotinin and was only significantly inhibited by the high-dose regimen. These data do indeed support our previous conclusion that the 2 million KIU pump prime dose of aprotinin improves hemostasis mainly by preservation of the platelet adhesive function. 8,9
However, the better result in hemostasis obtained with the high-dose regimen suggests that the hyperfibrinolysis observed after release of the aortic crossclamp might play an additional role. This hyperfibrinolysis could be caused by several factors that coincide with the release of the aortic crossclamp. First, normothermia is restored, resulting in normal antiprotease activity that is inhibited during hypothermia. 22 Cardiotomy suction is used more intensely and returns coagulation- and fibrinolysis-activated blood from the pericardial cavity, which results in systemic activation of the fibrinolytic system. 23 Reperfusion of the ischemic heart and lungs might be responsible for the increased tissue plasminogen activator activity observed in this period. 1 Moreover, a peak release of endotoxin (a potent stimulus for tissue plasminogen activator release), likely to be trapped in the thoracic duct during ventilatory arrest, appears in the systemic circulation. 24 Whatever the cause of the hyperfibrinolysis might be, complete inhibition by high-dose aprotinin but not by the low-dose regimen suggests that hyperfibrinolysis is a contributing factor in blood loss in cardiac operations. Improved hemostasis obtained with drugs that specifically inhibit fibrinolysis, such as 
-aminocaproic acid and tranexamic acid, confirms the contribution of fibrinolysis to the disturbed hemostasis in cardiac operations. 25,26
The mechanism of protection of the platelet adhesive function by aprotinin is still unclear. Many authors have suggested a primary role for plasmin in the reduced expression of the adhesive glycoprotein Ib receptors. 14,17,18 However, as our results show, fibrinolysis increases primarily after release of the aortic crossclamp whereas the reduced glycoprotein Ib expression was already observed at the beginning of bypass. 8,9 Moreover, low- and high-dose aprotinin were shown to be equally effective in preserving glycoprotein Ib expression, 9 but do not equally inhibit plasmin activity. The limited role of fibrinolysis is further evidenced by the observation that aprotinin also protects platelet function in simulated (in vitro) CPB in which plasmin concentrations do not increase because of the absence of endothelium-derived plasminogen activators. 27 These observations make a primary role of plasmin in damage to platelet function during CPB doubtful.
Another suggested mechanism of damage to platelet adhesive function during CPB is by thrombin generated through the intrinsic pathway on the surface of the extracorporeal circuit. 16,28 In the current study thrombin generation was measured with the markers F1+2 and TAT complex. A typical pattern was seen with increased thrombin activity after the onset of bypass and after release of the aortic crossclamp. This latter event coincides with increased cardiotomy suction. Tabuchi and associates 23 have observed that coagulation and fibrinolysis are activated in the pericardial cavity. Evidently this activation has to be explained by extrinsic activation. Because there is no indication that aprotinin has any effect on the extrinsic pathway this would explain the similar F1+2 and TAT levels in the groups.
The question is, however, whether small amounts of thrombin generated through the intrinsic pathway, which is inhibited by aprotinin, 24 can be distinguished in these measurements. Amplification of the activation of coagulation by factors V and VIII occurs on the phospholipid membranes of platelets. In case of extrinsic pathway activation this takes place locally in the platelet plug. With intrinsic pathway activation on the surface of the extracorporeal circuit, however, adjacent circulating platelets will provide the phospholipid surface to amplify thrombin generation. In these circumstances, even when only a small amount of thrombin is generated, a large proportion of the circulating platelets, which have a higher affinity for thrombin than fibrinogen, 29 can be affected by the generated thrombin. It remains questionable whether this small amount of thrombin generated, but with large consequences, can be distinguished against the background of extrinsically generated thrombin. Possibly this amount of intrinsically generated thrombin is reflected by the difference between the F1+2 levels in the aprotinin-treated groups and the placebo group.
Since the introduction of aprotinin many authors have expressed their fear of an increased prevalence of thromboembolic complications and graft thrombosis and reduced graft patency with the use of aprotinin as a result of an altered balance between thrombosis and fibrinolysis. 5,7,9 When the 2 million KIU schedule is used patients are likely to be protected from these complications inasmuch as it is administered only during bypass and full heparinization. With the 6 million KIU schedule, however, aprotinin is administered before and after bypass when there is no protective effect of heparin, but the clotting cascade is being activated because of intravascular lines, surgical trauma, de-endothelialized grafts, and suture materials. This might result in a hypercoagulable state in which thrombin production is unimpaired, but fibrinolysis is inhibited. 6 Moreover, aprotinin inhibits activated protein C, a natural anticoagulant that, after activation by thrombin, inhibits coagulation and promotes fibrinolysis. 5,30 The inhibition of this negative feedback mechanism of thrombin generation is likely to further increase the risk of clot formation. These factors, together with the unreliable ACT measurement, 31 make aprotinin a potentially dangerous drug.
The fear of reduced graft patencies or increased infarction rate has not been validated in follow-up studies. 7,10 On the other hand, in a recent study an increased prevalence of graft thrombosis was reported, 32 possibly caused by an inadequate heparinization protocol. Moreover, thrombus formation on pulmonary artery lines during aprotinin treatment has been observed, 33 and a study in a porcine model of arterial thrombosis showed an increased prevalence of arterial occlusion with aprotinin treatment. 34 To minimize the potential risk of these serious complications we believe aprotinin should not be administered when full heparinization is not in effect.
However, our results show that the low-dose regimen is less effective in blood loss reduction than the high-dose regimen and that this might be because of insufficient inhibition of hyperfibrinolysis after release of the aortic crossclamp. Thus it would be logical to increase the aprotinin levels by a second dose before release of the crossclamp. On the basis of the aprotinin levels in the low-dose group a single dose of 1 million KIU before release of the crossclamp can be expected to increase the levels to exceed 200 KIU/ml, which is sufficient for inhibition of plasmin. 1 With this modification the low-dose regimen is tailored to protect platelet adhesive function at the initial phase of CPB and to prevent the hyperfibrinolysis after release of the aortic crossclamp, while avoiding administration of aprotinin before heparinization and after neutralization of heparin.
Acknowledgments
We are indebted to R. J. Berckmans and M. C. L. Schaap for their skillful performance of laboratory assays and to J. J. P. Nauta, MSc, for his statistical advice.
Footnotes
From the Center for Cardiopulmonary Surgery, Department of Thoracic Surgery, Onze Lieve Vrouwe Gasthuis,a Amsterdam, and the Department of Clinical Chemistry, University Hospital,b Leiden, The Netherlands. 
References
and increases release of thromboxane B2 in cultured human umbilical vein endothelial cells. J THORAC CARDIOVASC SURG1992;104:654-8.2-plasmin-inhibitor complexes during cardiopulmonary bypass. Blood 1989;73:468-71.This article has been cited by other articles:
|
|
 |
|
  A. B. A. Vonk, M. I. Meesters, J. Schats, J. W. A. Romijn, E. K. Jansen, and C. Boer Removal of aprotinin from low-dose aprotinin/tranexamic acid antifibrinolytic therapy increases transfusion requirements in cardiothoracic surgery Interact CardioVasc Thorac Surg, February 1, 2011; 12(2): 135 - 140. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. D. McEvoy, S. T. Reeves, J. G. Reves, and F. G. Spinale Aprotinin in Cardiac Surgery: A Review of Conventional and Novel Mechanisms of Action Anesth. Analg., October 1, 2007; 105(4): 949 - 962. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. R. Brown, N. J.O. Birkmeyer, and G. T. O'Connor Meta-Analysis Comparing the Effectiveness and Adverse Outcomes of Antifibrinolytic Agents in Cardiac Surgery Circulation, June 5, 2007; 115(22): 2801 - 2813. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 The Society of Thoracic Surgeons Blood Conservatio, V. A. Ferraris, S. P. Ferraris, S. P. Saha, E. A. Hessel II, C. K. Haan, B. D. Royston, C. R. Bridges, R. S.D. Higgins, G. Despotis, et al. Perioperative Blood Transfusion and Blood Conservation in Cardiac Surgery: The Society of Thoracic Surgeons and The Society of Cardiovascular Anesthesiologists Clinical Practice Guideline Ann. Thorac. Surg., May 1, 2007; 83(5_Supplement): S27 - S86. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. Fergusson, K. C. Glass, B. Hutton, and S. Shapiro Randomized controlled trials of aprotinin in cardiac surgery: could clinical equipoise have stopped the bleeding? Clinical Trials, June 1, 2005; 2(3): 218 - 232. [Abstract] [PDF]
|
|
|
|
|
|
R. J. Frumento, C. M.N. O'Malley, and E. Bennett-Guerrero Stroke after cardiac surgery: a retrospective analysis of the effect of aprotinin dosing regimens Ann. Thorac. Surg., February 1, 2003; 75(2): 479 - 483. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. J. Munoz, N. J. O. Birkmeyer, J. D. Birkmeyer, G. T. O'Connor, and L. J. Dacey Is {epsilon}-Aminocaproic Acid as Effective as Aprotinin in Reducing Bleeding With Cardiac Surgery? : A Meta-Analysis Circulation, January 12, 1999; 99(1): 81 - 89. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. G.H. Speekenbrink, R. M. Bertina, F. Espana, C. R.H. Wildevuur, and L. Eijsman Activation of the protein C system during cardiopulmonary bypass with and without aprotinin Ann. Thorac. Surg., December 1, 1998; 66(6): 1998 - 2002. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |
