Secondary Logo

Journal Logo

Original Article

Preventing and managing postoperative bleeding in cardiac surgery by controlling disseminated intravascular coagulation

Szefner, J. MD

Author Information
European Journal of Anaesthesiology: June 2007 - Volume 24 - Issue - p 59-66
  • Free

Abstract

Introduction

Bleeding is a surgeon's constant concern, and how one treats it during an operation may make the difference between success and failure. Above all, one must determine whether it is surgical or biological, and if biological, one must be able to diagnose and treat it correctly.

It is clear that most bleeding in cardiovascular surgery, rather than being surgical, has a biological origin.

Abnormal biological bleeding can be due to:

  • Acquired defects in the formation of the platelet plug (often iatrogenic in origin)
  • A reduction in the levels and function of coagulation factors (also frequently iatrogenic in origin)
  • Inadequate heparinization, or the possibility of subsequent inadequate neutralization, or excessive protamine dose
  • Qualitative defects in the polymerization of fibrin
  • But above all, to disseminated intravascular coagulation (DIC) in its latter phases, as the result of increased fibrinolytic activity.

Accordingly, the follow-up and treatment of DIC is of the utmost importance in order to anticipate and slow down causes of its occurrence and prevent further deterioration of the hemostatic systems.

Biological bleeding, and in particular DIC, is a neglected, little understood phenomenon. DIC can be defined as a pathological process in which a generalized hyperactivation of the hemostatic systems occurs by one or more triggering pathways, causing widespread thrombin, fibrin and plasmin formation and subsequent lysis within the vascular system which often leads to microthrombosis and consumption of platelets, as well as of clotting and inhibitor factors.

DIC can be summarized with one word: Hypercoagulability; but this phenomenon is very difficult to delimit. If biological hypocoagulability has an evident clinical translation, bleeding or a bleeding tendency, hypercoagulability does not always result in thrombosis. This fact has always troubled the clinician and especially the surgeon, since there is no distinct parameter of hypercoagulability that can predict thrombosis.

When we analyze the different phases of DIC, we see that DIC biological hypocoagulability, manifested clinically as hemorrhage, is simply the translation of a state of overt and decompensated hypercoagulability.

In nearly all cases, DIC is a secondary response to a pre-existing primary pathology.

Frequently, the pathology is in fact the expression of cellular stress, which is reversible when a patient's condition improves, or irreversible when the stress results in the death of cells and a consequent partial or complete functional loss of tissue or of the organ of which they are a part, or even a patient's death if a vital organ is affected.

Cellular stress is the result of aggression. The aggressing agents can either be exogenous, such as microorganisms, their toxins and physical-chemical agents, or endogenous, such as qualitatively and quantitatively abnormal metabolic products (urea, uric acid, fatty acids, etc.). Moreover, the latter can be induced by the former in a chain reaction.

The aggressing agents may also be of mixed origin. This is the case with antibodies, which are humoral reactions to extraneous allergens (hetero-, iso- and auto-immune diseases). These aggressing agents may exert their harmful effects on cells in two ways: either directly by upsetting their metabolism, or indirectly by disrupting particularly receptive blood cells, i.e. platelets.

This is a necessarily frequent phenomenon, since the agents' entry pathway is usually blood and the first cells which they encounter are of course platelets. Accordingly, by exacerbating the platelets' fundamental properties (adhesion, releasing of coagulation factors, aggregation), they induce obstruction events or intravascular coagulation in the microcirculation, which indirectly affect organs' basic cells by provoking their ‘anoxia.’ If this local ischemia, whose extent is determined by the vessels affected, is not rapidly corrected, the consequence is necrosis.

The latter process can even be triggered by aggressing agents which are inoffensive for tissue cells, but not for platelets, because platelets, by virtue of their functions, are highly sensitive to numerous agents (catecholamines, fatty acids, collagen, antigen-antibody complexes, various enzymes, etc.). Moreover, the cytoplasm contents of all cells include variable quantities of thromboplastic effect substances, and are therefore procoagulant. Although these substances are not in themselves toxic, if they should succeed in entering into circulating blood (traumatism) or come from the cells themselves (hemolysis, leukocytes lysis), intravascular coagulation will then be triggered wherever there are favorable hemodynamic conditions (stasis, vasodilation, microcirculation, etc.). In this area, peptidases play an important role, since by accidentally penetrating into circulation they are likely to trigger the coagulation process in an extremely powerful way. If these events are often transitory and occur undetected, it is because organs are protected by strong defenses, in the form of their fibrinolytic potential, triggered by coagulation itself, which is the only means of assuring fast revascularisation. Furthermore, some organs, such as kidneys, which are particularly exposed to procoagulant agents, possess their own enzymatic defense system (urokinase), thereby reinforcing the protecting effect of the over-all fibrinolytic system.

Nevertheless, whether the defensive function is found to be deficient or has been exhausted by too frequent solicitation, the result is a dysfunction favoring the pro-coagulant action, thereby bringing on a circulatory arrest, an ischemia, necrosis and the loss of the concerned organ's potential function.

Most frequently, the DIC thrombophilic condition is related to multiple and different etiologies, which are mutually enhancing. In a state of acute infection, for example, thrombophilia can be induced at the same time by:

  • thromboplastic substances liberated by the lysed red blood cells,
  • the action of erythrocytic ADP on platelets,
  • reactional hyperfibrinogenemia,
  • microbic as well as leukocytic peptidases and cytokines (TNFα…), which are subsequently sustained and aggravated by the secondary stasis brought about by hypovolemia, eventually exacerbated by collapse or shock.

Thus the multiplicity of agents susceptible of altering coagulation, coupled with their interaction and mutual reinforcement, explain the frequency, usually undetected, of a pathological process which is common to a great number of situations reflecting extremely different etiologies, leading to an organic functional insufficiency, and often death.

This common denominator, the thrombophilic state, is in fact the source of serious complications in numerous illnesses. Accordingly, since effective therapeutic means are available to counter this threat, it is clear they should be neither ignored nor underestimated.

Therapeutic approach

Patients presenting such pathologies, as described earlier, suffer a strong and permanent aggression (continuous injury) at every level of the coagulation systems, which are always disturbed. Accordingly, if we want to prevent bleeding, we must imperatively control all the systems involved - i.e. the platelet system, the procoagulant system and its regulation, and the fibrinolytic system, because all these systems are tightly interwoven - and adapt the treatment to each individual.

Hemostasis, that is to say the spontaneous and physiological arrest of an « in vivo» hemorrhage, is the result of an unbalanced coagulolytic function tending towards hypercoagulability, but its effect is limited to the area in question, and it is harmonious, transitory and reversible.

A pathological intravascular coagulation (thrombosis) will develop from an analogous unbalanced situation, but it will be chaotic, extensive and often irreversible.

These two modifications, the first desirable, and the second dreaded, are initiated by the same mechanisms and are subject to the same influences. Only the triggering cause is different.

Any disturbance, therefore, initiates a reactional activity which can exceed its objective and constitute a pathological condition. This is the case with most fibrinolysis of whatever origin, acting on all levels of hemostasis, which follow an increase in coagulation potential. It is also the case with consumption hemorrhages which are established following an exhaustion of factors due to generalized coagulation in the microcirculation, consequently aggravated by a reactional fibrinolysis.

The result is a series of vicious circles. In order to address this situation effectively, these vicious circles must be broken at one or more points and treated with an apparently paradoxical therapy.

The platelet system

The platelet plays an important role in the triggering and maintenance of a thrombophilic state that can degenerate into DIC and thrombosis. This state must be controlled and corrected according to the different surgical and clinical conditions to which the patient and his platelet system are subjected.

Adhesion, the ‘release’ phenomenon, and hyperaggregability can be provoked by a great number of agents (ADP, collagen, catecholamines, fatty acids, antigen-antibody complexes, lactic acid, thrombin, peptidases, bio-materials, devices, flow changes, etc.). The initial effect is the mechanical obliteration of microvessels by blood clumps, quickly followed by a fibrin-forming process.

Moreover, endogen proteases (trypsin, leukocyte and tissue peptidases) or exogen proteases (microbic, insect and snake venoms), even in minute quantities, by accidentally penetrating in circulating blood can induce explosive hypercoagulability and exacerbate the platelet functions.

Among the secondary causes, hemodynamic disruptions are more frequent. Stasis, whatever its origin, brings together all the conditions favoring the formation of a thrombus. What is more, it is susceptible of independently bringing on a coagulation process by provoking a contact effect and by allowing platelets to adhere to the vessel walls and aggregate.

The metabolic or respiratory acidosis has a twofold effect on coagulation by reducing the inhibiting effect of ‘physiological’ and therapeutic Heparin as the pH drops, and by inducing platelet hyperadhesion.

Any increase in plasmatic proteins, in particular fibrinogen, affects hemodynamics (hyperviscosity) and represents one of the most important causes of structural hypercoagulability (raw material of a clot). Moreover, there is a poorly defined, yet clear, relation between fibrinogen and the activation of platelet functions.

In patients undergoing strong and permanent aggression, there is a basic discrepancy between the aggregation curves and those resulting from the determination of Platelet Factor 4 (PF4) and Beta Thromboglobulin (BTG), thus reflecting platelet hyperactivation. Very few platelets remain intact, explaining the hypoaggregability and possible hemorrhage. The content of the platelet is now external (PF4, BTG, thrombin), which increases the activation of other platelets and the formation of circulating aggregates, and prevents the appropriate use of Heparin.

Because the degranulation phenomenon is closely related to the change in the intra-platelet level of cyclic AMP, it is possible to avoid this phenomenon by increasing the beneficial level of intra-platelet cAMP by inhibiting phosphodiesterase with appropriate doses of Dipyridamole. The platelet then becomes less receptive to usual inductors, preventing the release of the platelet's secreted products which can generate endothelial changes and thus activate other platelets.

The procoagulant system and its regulation

The involvement of the procoagulant system and its regulation in almost all chronic or acute aggressions has a triple consequence:

  • In acute and subacute forms, this function can be responsible for severe diffuse hemorrhages or a local hemorrhage in a vital organ, by means of a consumption process which may or may not be associated with a reactional fibrinolysis;
  • The same clinical result can be directly obtained by a fibrinolysis without major consumption, the coagulant process having only triggered fibrinolysis, which in this case, dominates.
  • In chronic forms, far more frequent than one might expect, the pathogenic evolution is incomplete and the ‘consumption’ is often insufficient to bring about hemorrhages. The process is then limited to vascular obliteration inducing ischemia, necrosis lesions and the functional loss of the organ in question. Examples are organ rejection after transplantation, or more often, the acute renal insufficiency that complicates so many diverse etiological pathologies.

Concerning coagulation itself, the goal is to establish a balance between the different pathways of thrombin formation and the activity of the most powerful inhibitor, Antithrombin III (AT III).

Such a balance, we believe, can be best evaluated via thromboelastography (TEG) on recalcified whole blood (which includes the Thrombodynamic Potential Index, TPI -normal value with ‘steel’ cups and pins: 6-15; with ‘plastic’ cups and pins: 8-20), the only sensitized test that allows study of global coagulability and the different phases of coagulation. This test makes it possible to determine the kinetics, dynamics, syneresis and structural quality of the clot.

It is important to point out that evaluating only plasma is not sufficient (especially during the imbalance due to a hemodilution, as in cardiopulmonary bypass- CPB), no matter what the test, since the patient performs his hemostatic functions on whole blood, including thrombi. Thromboelastography allows us to assess overall hypercoagulability and to appreciate the intensity of an early, lytic evolving process.

Coagulation balance may also be evaluated by establishing the Antithrombin Potential Index (API), which makes it possible to appreciate the adaptation capacity of inhibition to global coagulability. The API is determined by subtracting AT III in serum from AT III in plasma (normal value: 35-45).

And finally, determining the Activated Factor X in serum (normal value: 10-20%) indicates the speed of thrombin formation.

To obtain a satisfactory balance, the most appropriate anticoagulant to date is undoubtedly Heparin. For CPB, we administer 120-200 IU/Kg of Heparin and we adapt doses during an operation to reach a Hemochron or activated clotting time (ACT) of around 450 ± 50 seconds.

The neutralization of anticoagulation with Protamine at 60% (≈2/3) of the total dose of Heparin injected during CPB is sufficient, because the relative excess of circulating Heparin left after the intervention is only apparent, since it is necessary to take into account Heparin administered before intervention and turnover during the operation.

After the administration of Protamine, blood recovers its coagulation potential, although in reality it is not normal. The patient is actually hypercoagulable due to surgery itself, the existence of thromboplastic material released during surgery, the circulatory arrests and stasis, anoxia, acidosis, etc. Huge amounts of thrombin are generated and subsequent fibrinogen is activated despite the presence of presumed adequate heparinization during CPB. We should therefore understand that the increased coagulant function potential of those patients is partially manifested by both the chronometric global coagulability (indicated by ‘conventional’ tests), and even more important, by structural coagulability which is not revealed by conventional tests and therefore remains undetected if it has not been possible to examine blood with thromboelastography. To be sure, it is essential to also examine the action of inhibitors, and not only their presence in plasma, as well as verify their capacity to adapt to the global coagulability of a patient.

The surgeon should therefore be alerted to a possible presence and likely imminent occurrence of DIC, especially in the presence of severe hemodynamic, metabolic, and hepatic perturbation, with probable sepsis, when a systemic inflammatory response syndrome is present, during surgical re-interventions, or when CPB has been especially long, making it necessary to administer Heparin in order to neutralize the thromboplastic material released and break the vicious circle of a self-sustained coagulation.

Following CPB, in order to maintain appropriate anticoagulation and in case of DIC, imbalances are corrected by low-dose Heparin (1000 to 5000 IU/day). Low dose Heparin has multiple sites of action, induces endothelial modulation and releases Tissue Factor Pathway Inhibitor (TFPI). Heparin therapy is adapted and monitored using thromboelastography to carry out Raby's transfer test on plasma. We study the TEG's constant « r » - that is to say, the kinetics of the clot formation - of a control and of a mixture patient-control, from which an index is obtained.

The Heparin therapy established in our protocol enables us to obtain an efficient antithrombinic effect and prophylactic effect. At the same time, it allows the inhibitory system to function adequately and permits enough thrombin formation to ensure satisfactory hemostasis, although it is insufficient to bring about a state of real hypercoagulability.

The fibrinolytic system

After having stabilized the platelet function and balanced coagulation, we must stop fibrinolysis.

If we examine a large caliber vessel (artery, vein) and a capillary of a patient, we observe that thrombosis is produced in the large vessels and not in the small. Paradoxically, small vessels are the site of a great deal more obliterations, because of their anatomical configuration encouraging stasis, platelet deposits and the coagulation phenomenon.

This seeming paradox is explained by the fact that the intensity of fibrinolytic activity is inversely proportional to the diameter of the blood vessel. In other words, the fibrinolytic reaction in the capillaries is more powerful because the contact surface between a clot and the vessel's wall is much greater.

Accordingly, at the level of microcirculation, we observe successive waves of clot formation and lysis which may lead to an excessive consumption of coagulation factors and a hemorrhage syndrome masking hypercoagulability. Whether it be ‘real’ (kinetic and dynamic hypercoagulability either associated or not with platelet hyperaggregability), or ‘masked’ (thus requiring Raby's transfer test, for example), it is essential to be able to recognize this situation and appreciate its intensity so as to determine the necessary preventive or curative treatment. In order to be effective, the curative therapy cannot be carried out by means of ‘standard’ posology, but rather by an adapted and personalized dosage for each particular case determined by biological controls.

Fibrinolysis is due to the activation of plasminogen once it has overcome the effect of the inhibitors of this action. Active plasmin protease thus formed is physiologically inhibited mainly by Alpha-2-antiplasmin (α2AP). This inhibition can lead to the system's exhaustion if the activation of plasminogen is too great or lasts too long.

Thus we see, on the one hand, the strong relationship between DIC and bleeding, and on the other hand, the need to look for DIC even if there is no bleeding.

DIC, Diagnostic and Therapeutic Approach

Bleeding is a dynamic phenomenon. The multi-system coagulation protocol which we have put in place in our department at La Pitié Hospital recognizes four distinct DIC phases:

  • Compensated thrombophilic phase.
  • Clot kinetic-structure dissociation phase.
  • Generalized hypocoagulability phase, with the ‘pinching’ phenomenon.
  • Secondary fibrinolysis phase.

Each phase has a different biological profile and requires a different therapeutic approach.

Phase 1 (Compensated Thrombophilic Phase). During the first phase, we can say that the « essential structure » appears to be intact: the factors and platelets are almost normal and Fibrinogen is normal or slightly increased. Nevertheless, TEG indicates structural and kinetic hypercoagulability, even though the inhibitor, AT III, is compensating for this excessive solicitation resulting from greatly accelerated thrombin formation.

Solicitation of the fibrinolytic system is either absent or extremely weak, the platelet function is relatively intact and there is no specific clinical profile, except for the cause triggering this phase.

What should we do at this phase? Little or nothing, other than closely follow the patient's evolution because if the cause or causes persist in the context of this situation, we may very quickly find ourselves facing later phases.

Phase 2 (Clot Kinetic-Structure Dissociation Phase). In the second phase, the factors are moderately diminished. Fibrinogen may have diminished in quantity and in quality but it also may have increased (in sepsis, for example), and reptilase time is longer. Reptilase time reflects FDP's activity even in the presence of Heparin.

TEG can best identify this phase. It reveals there is a dissociation between a rapidly forming clot (accelerated kinetics) that is consuming a great deal of raw materials, but which at the same time is of poor quality (‘soft’ clot) and which can lyse easily. Plasma AT III is strongly solicited although it can still keep up and compensate, providing a consequent diminution of serum AT III and a normal or slightly diminished API.

We also observe a drop in the number of platelets and in their activity, as well as an early triggering of the fibrinolytic system observed by a decrease in plasminogen, together with circulating plasmin and thus a drop in α2AP.

Clinically, we often observe the diffuse presence of isolated or converging petechiae, of hematomas, etc. This phase is very frequent in the immediate postoperative period.

To break the vicious circle (thrombin formation - plasmin formation - lysis), we administer high doses of Dipyridamole early on in order to stabilize the platelet function, and low doses of Heparin so as to slow down thrombin formation, because acceleration of conditions in the second phase will inevitably lead to the next phase.

Phase 3 (Generalized Hypocoagulability Phase). As reflected in TEG, the third phase presents a clear picture of hypocoagulability, including a dramatic drop in factors and virtual absence of the platelet function. Fibrinogen, however, is only slightly affected and Raby's Transfer Test is positive, indicating circulation of great quantities of thromboplastic material, thus triggering activation of the fibrinolytic system.

This is clinically translated by an aggravation of the previous phase. In addition, there is a clear tendency to suffer frequent and profuse hemorrhages at oozing points and in the mucosae. Treatment administered in phase 2 should be maintained and consumed factors should be replaced with fresh frozen plasma (FFP), AT III and/or fresh whole blood.

The ‘pinching’ phenomenon (hypocoagulability - reflected by a TPI < 6 from TEG, associated with a decrease of plasma AT III activity - either because there is insufficient production or because there is excessive consumption - and a rise in serum AT III activity, giving a significantly reduced API - < 20) is almost always detectable in this phase and reflects the inhibitor's inability to compensate for growing consumption of procoagulant factors. This is translated biologically in TEG by hypocoagulability (with, of course, underlying hypercoagulability) and clinically by hemorrhage.

Clearly, the ‘pinching’ phenomenon is a valuable prognostic tool meriting further study to better understand hypercoagulability.

Phase 4 (Secondary Fibrinolysis Phase). In the final phase of DIC, the massive and diffuse hemorrhage is catastrophic, reflecting a breakdown of all systems. Hypocoagulability appreciated by thromboelastography is accompanied by an almost complete absence of procoagulant material and inhibitory activity, together with totally ineffective platelets. At the same time, and this is of great importance, we observe that the fibrinolytic system is strongly solicited, provoking the disappearance of Alpha-2-Antiplasmin. As a result, plasmin is able to circulate freely and is self-maintaining, resulting in a negative Raby's Transfer Test.

Accordingly, to the treatment applied in the previous phase, we must add Aprotinin at a dose of 125,000 or 250,000 KIU administered intravenously (repeated, if necessary, at half-hour intervals), followed by a drip of 500 KIU per minute. We should also point out that Aprotinin is administered to some patients in phase 3, if there is a strong drop in α2AP. In addition, there should be a vigorous replenishment - although modulated - of AT III, FFP, etc., in phase 4; in short, of everything which is lacking.

Discussion

Clear evidence that DIC bleeding can be controlled is provided by a study carried out by Glauber and Ferrazzi in a series of cardiac assists over time in which they compare ‘conventional’ treatment of bleeding with treatment based on our multi-system protocol. In their experience, bleeding and the need for re-operations were greatly reduced thanks to the protocol, with a consequent increase in the survival rate.

It is no exaggeration to say that the conventional approach to bleeding is to ‘fill up the reservoir,’ which results in ever greater consumption of materials. Our approach adopts a radically different strategy. First, guided by the biological diagnosis, we ascertain the patient's DIC phase. Second, we stop the bleeding by treating the biological disturbances of the patient and only afterwards replenish the material that is lacking.

Strict control based on appropriate tests is essential before the operation, at the end of CPB and daily following the operation. More assessments may be necessary in case of complication. The protocol thus makes possible a significant drop in mean closure time, total thoracic drainage, re-exploration and blood derivatives transfusion.

Aprotinin is usually applied in large doses during CPB because of its effect on fibrinolysis and its presumed effect on the platelet function. In fact, Aprotinin has no direct influence on the platelet function. Accordingly, we stop fibrinolysis with Aprotinin, not in large doses systematically during CPB as is usually done, since such an approach is usually unnecessary and can be dangerous, but at the end of the operation in modulated doses based on biological criteria. It should be added that administration of the drug is frequently prolonged at the same doses in the ICU.

Aside from the somewhat excessive procoagulant effects of Aprotinin observed in vascular surgery or when controlling CPB, it is worthwhile to draw attention here to bleeding that ‘resumes’ shortly after the administration of high doses of this molecule, in particular when accompanied by FFP transfusions.

Among other factors, FFP provides additional α2AP (although ‘weakened’ by previous freezing, conservation and unfreezing), which then ‘competes’ with Aprotinin, often resulting in so-called ‘unexplained’ bleeding.

We believe that the adapted doses recommended in our protocol are sufficient and effective, while avoiding unnecessary complications.

The Antithrombin III role

We consider the use of AT III is indicated when there is:

  1. An acquired deficit in AT III of <60% following CPB
  2. An acquired deficit in AT III of <70% following CPB
    1. In the presence of increased thromboembolic and/or hemorrhagic risk factors, such as: re-interventions, cardiac valves, transplants (cardiac, pulmonary, hepatic, combined), circulatory assist devices, total artificial heart, age > 70, diabetes, long CPB (> 150′), hepatic insufficiency (primitive or cardiogenic)
    2. In DIC following cardiac surgery
  3. Long-term UFH treatment in the context of a persistent post surgery deficit in AT III
  4. Pre- or per-CPB heparin resistance

It is our belief that ‘conventional’ anticoagulant, platelet and antifibrinolytic therapy, however well intentioned, is based on several outdated, mistaken concepts accepted without question by generations of practitioners. In reality, such an approach does not take into account sufficiently three realities: the physiology of coagulation, the pathology of the thromboembolic phenomenon and the pharmacology of the drugs prescribed.

Conclusions

Bleeding is not a fatality. Quite the contrary, it must be treated through a multi-system approach, and as early as possible, preferably at a biological phase before clinical bleeding. In addition, early monitoring and treatment adapted to each patient makes it possible to better follow a patient's evolution and helps reduce the occurrence of later thromboembolic events.

Selected Bibliography

1. Bellon JL, Szefner J, Cabrol C. Coagulation et coeur artificiel, 1st ed. Paris: Masson, 1989.
    2. Bonaros N, Mueller MR, Salat A et al. Extensive coagulation monitoring in patients after implantation of the MicroMed DeBakey continuous flow axial pump. ASAIO Journal 2004; 50: 424-431.
      3. Bukhari EA, Krukenkamp IB, Burns PG et al. Does Aprotinin increase the myocardial damage in the setting of ischemia and preconditioning? Ann Thorac Surg 1995; 60: 307-310.
        4. Cabrol C, Pavie A, Szefner J et al. Management of surgical bleeding in cardiopulmonary bypass surgery In: Pifarré R (ed). Management of bleeding in cardiovascular surgery. Philadelphia: Hanley & Belfus, 2000: pp 469-474.
          5. Copeland J, Arabia F, Tsau P et al. Total artificial hearts bridge to transplantation. Cardiol Clin 2003; 21: 105-117.
            6. Copeland J, Szefner J. Anticoagulants and the artificial heart In: Lewis T, Graham TR (eds). Mechanical circulatory support. London: Edward Arnold, 1995: pp 306-311.
              7. Dewald O, Fischlein T, Vetter HO et al. Platelet morphology in patients with mechanical circulatory support. Eur J Cardiothorac Surg 1997; 12: 634-641.
                8. Drews T, Loebe M, Szefner J et al. Assistance biventriculaire de longue durée avec le système d'assistance pneumatique biventriculaire externe Berlin-Heart. J Chir Thorac Cardiovasc 2000; 3: 129-136.
                  9. Gando S, Kameue T, Nanzaki S, Nakanishi Y. Disseminated intravascular coagulation is a frequent complication of systemic inflammatory response syndrome. Thromb Haemost 1996; 75: 224-228.
                    10. Glauber M, Szefner J, Senni M et al. Reduction of haemorrhagic complications during mechanically assisted circulation with the use of a multi-system anticoagulation protocol. Int J Artif Organs 1995; 18: 649-655.
                      11. Harker L. Bleeding after cardiopulmonary bypass. N Engl J Med 1986; 314: 1446-1448.
                        12. Hathaway WE, Goodnight Jr SH. Disseminated intravascular coagulation In: Hathaway WE, Goodnight Jr SH (eds). Disorders of hemostasis and thrombosis: a clinical guide. New York: McGraw-Hill Inc., 1993: pp 219-229.
                          13. Koster A, Loebe M, Hansen R et al. Alterations in coagulation after implantation of a pulsatile Novacor LVAD and the axial flow MicroMed DeBakey LVAD. Ann Thorac Surg 2000; 70: 533-537.
                            14. Lemmer Jr JH, Despotis GJ. Antithrombin III concentrate to treat heparin resistance in patients undergoing cardiac surgery. J Thorac Cardiovasc Surg 2002; 123: 213-217.
                              15. Leprince P, Pavie A, Szefner J et al. Le cœur artificiel total Cardiowest, à propos de 29 cas. Arch Mal Cœur et Vx 1998; 91: 397-404.
                                16. Livingston ER, Fischer CA, Bibidakis EJ et al. Increased activation of the coagulation and fibrinolytic systems leads to hemorrhagic complications during left ventricular assist implantation. Circulation 1996; 94(suppl II): 227-234.
                                  17. Mohr R, Goor DA, Lavee J. Fresh whole blood In: Mohr R, Goor DA, Lavee J (eds). Management of bleeding after open heart surgery. Heidelberg: Springer, 1997: pp 111-121.
                                    18. Murphy PJ, Connery C, Hicks GL et al. Homologous blood transfusion as a risk factor for postoperative infection after coronary artery bypass graft operations. J Thorac Cardiovasc Surg 1992; 104: 1092-1099.
                                      19. Pavie A, Szefner J, Leger P et al. Preventing, minimizing and managing postoperative bleeding. Ann Thorac Surg 1999; 68: 705-710.
                                        20. Quaini E, Pavie A, Chieco S, Mambrito B, and the Registry Committee. The Concerted Action “Heart” European registry on clinical application of mechanical circulatory support systems: bridge to transplant. Eur J Cardiothorac Surg 1997; 11: 182-188.
                                          21. Raby C. Coagulations intravasculaires disséminées et localisées, 2nd ed. Paris: Masson, 1974.
                                            22. Ranucci M, Ditta A, Boncilli A et al. Determinants of antithrombin consumption in cardiac operations requiring cardiopulmonary bypass. Perfusion 2004; 19: 47-52.
                                              23. Ranucci M, Frigiola A Menicanti L et al. Postoperative antithrombin levels and outcome in cardiac operations. Crit Care Med 2005; 33: 355-360.
                                                24. Salsberg S, Lachat M, Zünd G et al. Left ventricular assist device as bridge to heart transplantation - lessons learned with the MicroMed DeBakey axial blood flow pump. Eur J Cardiothorac Surg 2003; 24: 113-118.
                                                  25. Spanier T, Oz M, Levin H et al. Activation of coagulation and fibrinolytic pathways in patients with left ventricular assist devices. J Thorac Cardiovasc Surg 1996; 112: 1090-1097.
                                                    26. Szefner J, Cabrol C. Control and treatment of hemostasis in patients with a total artificial heart: The experience of La Pitié In: Pifarre R (ed). Anticoagulation, Hemostasis, and Blood Preservation in Cardiovascular Surgery. Philadelphia: Hanley & Belfus, 1993: pp 237-264.
                                                      27. Szefner J, Pavie A, Leger P et al. Supervision and treatment of hemostasis in 82 patients implanted with a total artificial heart at La Pitié hospital. Thromb Haemost 1995; 73: 1043.
                                                        28. Szefner J. Apport de l'étude de la fonction plaquettaire et de la cinétique coagulo-lytique dans le contrôle et le traitement des transplantés cardiaques In: Laboratoires Sandoz (ed). Transplantation Cardiaque. France: Rueil-Malmaison, 1989: pp 80-87.
                                                          29. Szefner J. Control and treatment of hemostasis in cardiovascular surgery. The experience of La Pitié Hospital with patients on total artificial heart. Int J Artif Organs 1995; 18: 633-648.
                                                            30. Teoh KH, Christakis GT, Weisel RD et al. Dipyridamole preserved platelet and reduced blood loss after cardiopulmonary bypass. J Thorac Cardiovasc Surg 1988; 96: 332-341.
                                                              31. Vinazzer HA. Antithrombin III in shock and disseminated intravascular coagulation. Clin Appl Thrombosis/Hemostasis 1995; 1: 62-65.
                                                                32. Vitali E, Lanfranconi M, Ribera E et al. Successful experience in bridging patients to heart transplantation with the MicroMed DeBakey ventricular assist device. Ann Thorac Surg 2003; 75: 1200-1204.
                                                                  33. Wahba A, Black G, Koksch M et al. Aprotinin has no effect on platelet activation and adhesion during cardiopulmonary bypass. Thromb Haemost 1996; 5: 844-848.
                                                                    34. Wegner J, DiNardo J, Arabia F et al. Blood loss and transfusion requirements in patients implanted with a mechanical circulatory support device undergoing cardiac transplantation. J Heart Lung Transplant 2000; 19: 504-506.
                                                                      Keywords:

                                                                      Bleeding; Cardiac surgery; DIC; Assist Devices; Thromboelastography; Antithrombin III

                                                                      © 2007 European Society of Anaesthesiology