Cardiac surgical patients are at great risk for bleeding due to multiple factors. Many of these patients have underlying disease states that potentially may add to the risk for impaired haemostasis. Pharmacological therapy for cardiovascular disease, specifically anticoagulants and antiplatelet agents, can significantly contribute to these acquired causes of impaired haemostasis and bleeding. The additional effects of high-dose heparin anticoagulation for extracorporeal circulation, fibrinolysis, platelet activation associated with cardiopulmonary bypass, hypothermia and dilutional changes are some of the multiple factors that contribute to an acquired coagulopathy. Although pre-existing haematological diseases can further contribute to bleeding in cardiac surgical patients, most of the acquired haemostatic disorders are relatively uncommon. One of the critical haematological issues that significantly influences perioperative management in cardiac surgical patients includes heparin-induced thrombocytopenia (HIT). Other major issues related to hypercoagulability are also a potential issue that will be covered in understanding HIT and have been reviewed elsewhere.
Acquired haemostatic disorders due to pharmacological therapy are more common in cardiac surgical patients. A small percentage of patients (approximately 1%) may present with one or more of the common congenital bleeding disorders, most often haemophilia, von Willebrand’s disease (vWD) and/or inherited qualitative platelet defects. Although other disorders may exist in patients, focus will be on these common congenital haemostatic defects. Critical to patient management is the evolution of purified and recombinant replacement factors. The haemostatic factors administered for managing cardiac surgical patients with haemophilia undergoing surgery are recombinant or plasma-derived factor VIII for haemophilia A or factor IX for haemophilia B, von Willebrand factor (vWF) concentrates for patients with von Willebrand’s disease, and different therapies for patients with platelet defects including platelet transfusions.
Haemophilia is an inherited X-linked recessive disorder, with an incidence of ~1 in 5000 male births for haemophilia A (factor VIII deficiency) and 1 in 25,000 male births for haemophilia B (factor IX deficiency). Haemophilia patients develop bleeding into their joints, probably due to the extensive presence of glycosaminoglycans in the synovial surfaces and the lack of tissue factor. Both forms of haemophilia are characterised by low levels of factors VIII or IX. Patients with haemophilia routinely require orthopaedic surgery because of repeated bleeding into their joints. They may also undergo cardiac surgery. There are specific guidelines developed in the haemophilia literature, and haematological management should include consultations with a consultant haematologist for specific timing and maintenance of factor replacement therapy. Depending on the factor levels, factor concentrate replacements should be administered within 10–20 minutes of starting a procedure and levels followed perioperatively to further determine specific factor replacement requirements for maintaining levels in the first few perioperative days.
A critical haemostatic protein is vWF, which is present in the circulation as a multimeric macromolecule. Its haemostatic function facilitates platelet adhesion to damaged blood vessels, and has critical importance facilitating platelet aggregation. The macromolecule functions as a carrier protein to factor VIII, and circulates complex to this coagulation factor. The actual von Willebrand’s disease occurs in an estimated 1% of patients. The most common form of the disease is type 1, where vWF levels are 10% to 40%. Type 2 is a more complex qualitative vWF defect where the molecule has a decreased ability to bind factor VIII (type 2N), or increased (type 2B) or decreased (types 2A and 2M) ability to facilitate platelet-dependent vWF functions. Type 3 is the most severe form of the disease where vWF levels are <10% and factor VIII levels are also low. In addition, patients with severe aortic stenosis and ventricular assist devices develop an acquired form of the disease due to mechanical injury and disruption of the larger multimeric form. In patients with type 1 disease, desmopressin ~0.3 μg/kg is usually administered intravenously. However, in type 2 and type 3 vWD patients, vWF concentrates are the therapy of choice but should be administered after separation from cardiopulmonary bypass (CPB).
Although few clinical data are available, Bhave et al. reported one of the larger series of 17 patients undergoing cardiac surgery with CPB that included 13 patients with haemophilia A, one symptomatic haemophilia A carrier, one with haemophilia B and two with vWD. The cardiac surgical procedures included ten coronary revascularisations, two aortic valve replacements, two mitral valve repairs, two aortic root replacements, and one combined aortic valve replacement and coronary revascularisation. All procedures were managed by factor repletion to maintain normal levels. Two patients were re-explored for bleeding at 1 and 20 days postoperatively.
Other Less Common Inherited Bleeding Disorders
There is a paucity of information beyond case reports available for managing many of the other less common coagulation factor deficiencies. However, purified factor concentrates are available for most of the additional rare factor deficiency states. For many of these patients, fresh frozen plasma/plasma and other factors including factor concentrates may be important during management. Fortunately, many of these patients come previously diagnosed with their specific coagulation deficiency. Factor XII deficiency has important implications for cardiac surgery because patients present with a markedly prolonged partial thromboplastin time and activated clotting time, and can complicate managing anticoagulation with heparin for CPB. In patients with factor XII deficiency, fixed dose heparin should be administered based on time and/or duration of CPB and/or heparin monitoring using heparin-protamine titrations if available.
Beyond the common problem of acquired platelet dysfunction due to the use of P2Y12 receptor antagonists (e.g. clopidogrel, prasugrel and ticagrelor), rare forms of congenital platelet dysfunction include Glanzmann’s thrombasthenia and Bernard Soulier disease. Fortunately, most patients with this disease have a long history of easy bruisability, bleeding and/or family history. There are reports of these patients undergoing cardiac surgery. In patients with Glanzmann’s thrombasthenia, there is a defect on the platelet surface for the fibrinogen receptor GPIIb/IIIa that prevents platelet aggregation, which is a critical component of clot formation. Fortunately, patients may have a long history of bruising, mucosal bleeding and other soft-tissue bleeding which should make the clinician suspicious of such a disorder. Of note is that recombinant factor VIIa is approved for treating bleeding in this patient population. Another rare aforementioned syndrome is Bernard Soulier disease, where patients have thrombocytopenia and a reduced ability of the platelets to adhere to the damaged blood vessel owing to abnormality of the GPIb/V/IX receptor on the platelets, therefore preventing appropriate vWF binding, which is another critical step in platelet adhesion and activation. Other additional platelet abnormalities include storage pool disorders that may manifest in variable ways and are another rare cause of platelet dysfunction. Again, acquired platelet dysfunction due to the use of antiplatelet agents is a far greater issue in cardiac surgical patients.
‘Adequate anticoagulation’ for CPB is traditionally determined by the activated clotting time (ACT), a whole blood haemostatic activation test that is influenced by multiple causes. Heparin resistance has been frequently reported in the literature, often due to a variety of causes. The definition commonly used for heparin resistance is ACT <480 seconds following addition of 500 U/kg of heparin. Unfortunately, the terminology we use for heparin resistance is a misnomer, because actual resistance does not occur; rather there is an alteration in heparin dose responsiveness and an alteration of the slope of the heparin response curve. The ACT actually increases following heparin administration, but not in a linear fashion.
Multiple factors are responsible for producing altered heparin dose responses, in addition to antithrombin, including preoperative heparin use, high and low platelet counts (both thrombocytosis and thrombocytopenia), altered platelet function and use of other anticoagulants. The association between decreased preoperative antithrombin and heparin resistance is complicated by multiple factors that affect the ACT in addition to antithrombin levels. Ranucci et al. suggested a 10% predictive chance of a patient developing altered heparin dose responsiveness/heparin resistance. Antithrombin (also called antithrombin III) is a serine protease inhibitor that circulates in the plasma at 80–110% assuming normal levels, approximately 2.4 µmol. Antithrombin is associated with altered heparin responsiveness due to the fact that heparin requires the cofactor of antithrombin to inhibit factors Xa and thrombin (IIa). Antithrombin is activated by binding to heparin via the pentasaccharide sequence of heparin; the relative chain length of the sequence controls the binding selectively to factor Xa and/or IIa. Multiple investigators have noted that previous heparin administration decreases antithrombin levels, and produces alterations in heparin dose responsiveness.
Multiple reports note that antithrombin improves intraoperative anticoagulation, improves perioperative haemostasis and reduces biochemical markers of coagulation. Antithrombin administration to correct decreased levels of antithrombin would potentially be of benefit to all patients, as antithrombin levels fall to <50% during CPB. Although Linden et al. suggest therapeutic antithrombin is not of added benefit in heparin resistance because patients do not exhibit decreased antithrombin concentration compared with heparin responsive patients, this is not the case. Antithrombin administration will consistently increase ACT responsiveness in vitro and in vivo as previously reported.
Heparin-induced thrombocytopenia (HIT) is a prothrombotic disease where an anticoagulant has the potential to produce a procoagulant effect. The pathophysiology is due to an antibody that develops following heparin administration to a specific alpha granule protein released from platelets called platelet factor 4 (PF4). When PF4 is released, it fuses with the platelet membrane to form a new epitope/antigen, and as a result immunoglobulin (Ig) G develops that binds to PF4 complexes on platelets, causing activation, sequestration and microparticle formation, all of which have profound prothrombotic effects. Patients following cardiac surgery and CPB are at an increased risk of producing the antibodies which may occur in up to 50% of patients although potentially at low levels. The presence and the level of the antibodies are associated with an increased risk for adverse events, and occur in approximately 1–5% of patients who receive unfractionated heparin, although cardiac transplant patients may be at an even higher risk for HIT (~11%).
HIT should be suspected if the platelet count drops by 50%, or new thrombosis occurs in a patient 5 to 14 days after the start of heparin therapy. The 4T score (thrombocytopenia, timing of platelet count fall, thrombosis or other sequelae, and other causes for thrombocytopenia) has also been used to determine the probability of HIT and is thought to correlate well with antibody formation. However, in cardiac surgical patients in the perioperative setting, thrombocytopenia is a common problem and is due to multiple causes including dilutional changes post CPB, intra-aortic balloon pump presence and mechanical destruction, sepsis or other drug-induced causes, especially IIb/IIIa inhibitors. Following cardiac surgery and CPB, HIT may present in a biphasic pattern. Several days postoperatively, the platelet count begins to normalise after CPB-related thrombocytopenia, followed by a decreasing platelet count >4 days postoperatively. Following heparin administration intraoperatively, an acute hypersensitivity response may occur due to pre-existing IgG levels, with hypotension, other systemic responses and acute thrombocytopenia. HIT can also occur weeks after heparin exposure and is termed ‘delayed-onset HIT’ but this is less common.
Although suspicion of HIT is of critical importance to stop use of heparin and facilitate another anticoagulant, laboratory testing is an important adjunct to making the correct diagnosis. The standard testing includes antibody determination for screening and functional assay for definitive diagnosis. Antibody testing includes a standard enzyme-linked immunosorbent assay (ELISA) that detects either polyclonal antibodies or IgG specific antibodies to complexes of PF4 and heparin. ELISA results are commonly reported to clinicians as either positive or negative; specific information regarding optical density (OD) should also be obtained to facilitate clinical decisions regarding management and/or further workup. Based on the time of sampling, ‘high-titer negative’ ELISA results have the potential to become positive several days later if retested. More importantly, higher optical density results are associated with increased probability for true HIT, based on published data that will be discussed later. The serotonin release assay (SRA) and platelet aggregation testing are functional assays that are only undertaken in specialised laboratories.
Warkentin et al. reported a study that is important for clinical management and is based on antibody testing. In this study, the magnitude of positive ELISA results, as determined by optical density (OD) units, was correlated with functional platelet serotonin-release assay (SRA) results, the gold standard for determining whether a patient is indeed HIT positive. They reported that if the results were weakly positive, 0.40–1.00 OD units, the ELISA had a low probability of 5% or less for a positive SRA. However, the risk increased to ~90% with an OD of >2.00 units. Their study evaluated 1553 patient sera and reported that for every increase of 0.50 OD units in the ELISA-IgG, the risk of a positive SRA increased by OR = 6.39, and for every increase of 1.00 OD units in the EIA-IgG, the risk increased by OR = 40.81.
Postoperatively, in patients with clinical suspicion of HIT or confirmed HIT, stopping heparin and starting an alternative anticoagulant, most commonly the direct thrombin inhibitor bivalirudin or argatroban, are mainstays of therapy. The alternative danaparoid is not available in most countries. Management for CPB will be considered separately. As previously mentioned, the 4T score has been suggested to determine the relative risk of true HIT. Low molecular weight heparin should not be used because of potential cross-reactivity with antibodies. Anticoagulation with warfarin should be avoided based on guidelines until platelet count recovers. Heparin should be avoided, if possible, at least as long as heparin-PF4 antibody testing is positive.
For patients requiring cardiac surgery and CPB, although alternative anticoagulants can be used, heparin is still the best agent due to its acute reversibility. For patients with current HIT who require cardiac surgery, if elective, the surgery should be delayed until heparin-PF4 antibodies are negative. Alternatively, plasmapheresis has been recommended and used successfully to manage patients. However, alternative anticoagulation for CPB is best performed with bivalirudin, based on all of the reported data.