This chapter provides an overview of the coagulation system with emphasis on the impact of specific disorders and tests during the preoperative period. The body's hemostatic system could be compared with a lever that is in constant balance on a fulcrum. At one end is the physiological need to maintain blood in the fluid state, allowing unimpeded circulation of oxygen and nutrients to the tissues. At the other end is the need to restrict the outflow of blood should damage to the vasculature occur. This delicate balance exists as the result of a complex interaction among the three major components of the hemostatic system: the vascular compartment, circulating blood platelets, and coagulation proteins. The system is fine-tuned by the constant action of numerous naturally occurring inhibitors that modulate the body's response to minor vessel injury and maintain normal blood flow.
Overview of Coagulation
The Vessel Wall
The endothelium plays a major role in regulating membrane permeability, lipid transport, vasomotor tone, inflammation, vascular wall structure, and coagulation. The major components relevant to hemostasis are the intima lining, which is the point of contact between flowing blood and the vessel, and the muscular media, which is under the control of the autonomic nervous system. The intact vascular endothelium usually presents a thromboresistant surface. Several mechanisms are involved 1 : it dilates the vessel by producing several vasodilators, inhibits platelets and leukocyte activation, and has anticoagulant properties.
Of the vasodilators known today, nitric oxide (NO) seems to have an increasingly important role in maintaining the patency of the blood flow. Since the discovery in 1980 of an endothelium-derived relaxing factor 2 and the further characterization of this compound as NO in 1987, 3 it has been shown that NO is produced not only by endothelial cells but also by most tissues in the body. The main mechanism of NO action is the activation of soluble guanylate cyclase and subsequent production of cyclic guanosine monophosphate (cGMP). 4,5 cGMP has direct effects on protein kinases and phosphodiesterases. These kinases mediate a cGMP-induced decrease in calcium in the vascular smooth muscle, leading to a decrease in vessel tone. NO also has an effect on platelet function mediated via cGMP 6 : it inhibits platelet aggregation, secretion, adhesion, and binding to fibrinogen via the GPIIb/IIIa receptor.
Another substance released by the endothelium, prostacyclin (PGI2), has two main biological actions: to inhibit platelet aggregation and to serve as a powerful vasodilator. Both of these actions are mediated by cAMP. PGI2 secretion in humans is quite low, 7 making it unlikely that sufficient quantities are available for constant inhibition of platelet aggregation.
Although NO and PGI2 are important relaxing factors, they cannot explain all endothelium-related relaxation. In 1984, Bolton and colleagues 8 demonstrated that a muscarinic agonist elicited endothelium-dependent hyperpolarization and relaxation of vascular smooth muscle. This phenomenon seems to be at least partially resistant to inhibitors of NO and cyclooxygenase; therefore, the existence of an endothelium-derived hyperpolarazing factor has been postulated, and this factor would induce relaxation by opening potassium channels on smooth muscle cells. Current candidates for this factor are carbon monoxide and hydrogen peroxide. 9,10 The endothelial cell also secretes ectozyme, an enzyme that catalyzes the transformation of adenosine diphosphate (ADP) to adenosine monophosphate, which is a potent platelet inhibitor. 11
Another system that the endothelium possesses to suppress the activation of coagulation at the microvascular level is the expression of a tissue factor (TF) pathway inhibitor (TFPI), which binds to activated factor X (Xa). TFPI directly inactivates factor Xa and produces feedback inhibition of the factor VIIa/TF catalytic complex. Animal studies established that TFPI functions as a natural anticoagulant that protects rabbits from intravascular coagulation, triggered by the exposure of blood to small amounts of TF. 12
As we have seen, the endothelium functions as an anticoagulant surface. It does, however, possess constricting factors that include endothelin, thromboxane A2, and angiotensin II.
Under normal conditions, circulating cells such as leukocytes and platelets do not adhere to the endothelium because they circulate in the most central place of the vessel lumen and because adhesion molecules and their corresponding ligands are not expressed on the surface of platelets, leukocytes, and endothelial cells. Loss of the endothelial surface promotes platelet adhesion. Underneath the endothelium lies the intima, which is rich in connective tissue protein, capable of activating platelets to adhere to the vessel surface and subsequently aggregate. Two molecules are critical for this process: von Willebrand factor (VWF) and collagen. VWF acts as an intercellular glue that attaches platelets to the exposed subendothelium of the small vessels via the platelet receptor glycoprotein Ib. Consequently, platelets tumble over the subendothelium rather than flow by, but this is clearly incapable of acting as a barrier against the escape of blood under systemic pressure. A series of events now begin. 13 Platelets are “activated,” which means that they lose their discoid shape and start forming pseudopods, exposing procoagulant receptors (Table 1) and releasing the content of their granules. Alpha granules contain proteins such as platelet factor IV, fibrinogen, factor V, and VWF. The dense or δ granules contain adenosine triphosphate (ATP) ADP, calcium, and serotonin. The importance of these granules is manifested in the severe bleeding seen in congenital dense granules deficiencies, such as Hermansky-Pudlak syndrome. 14 Finally, all these events lead to a more stable platelet matrix, recruitment of more platelets, vasoconstriction, increased fibrin formation, and decreased fibrinolysis.
The Coagulation Cascade
While this platelet response takes place, the intima simultaneously initiates activity involving the plasma proteins of the coagulation cascade. Most of the coagulation proteins are synthesized in the liver and are thus dependent on intact hepatic activity for normal synthesis and function. In addition, factor VIII (antihemophilic factor) is manufactured in extrahepatic sites (Table 2). Classically, fibrin formation has been understood to be the result of two separate pathways of activation. For conceptual and in vitro laboratory purposes, we have been taught that there is an intrinsic pathway and an extrinsic pathway, ending in a final common pathway leading to fibrin formation. There is a lot of interaction between both processes; and, in general, what is called a coagulation cascade is a series of protein or zymogen conversions to active proteins or enzymes. Factor activation proceeds in a sequential fashion, with each factor acting as a substrate in an enzymatic reaction catalyzed by the previous factor in the sequence. Enzymatic cleavage of a protein fragment activates the enzymes, which are called “serine proteases” because the active site for its protease activity is a serine aminoacid residue (Figure 1).
The principal initiating pathway of blood coagulation in vivo is the extrinsic system, which uses components from both blood and vascular elements. This pathway begins with the release from damaged endothelium cells of a complex protein called tissue factor (TF). TF is not normally in contact with any plasma-clotting factor. After endothelial injury, TF activates coagulation factor VII. Once this factor is activated, it combines with calcium, resulting in the activation of factor X and, ultimately, in the generation of fibrin.
The intrinsic system runs in close interaction with the extrinsic system and can be considered as coagulation initiated exclusively by components within the vascular system. It involves more protein-protein interactions and proceeds at a slower rate than the extrinsic pathway. The intrinsic pathway begins with the exposure of plasma to the subendothelial intima, which causes structural rearrangement of factor XII. Factor XII then causes, in a sequential and amplifying manner, the conversion of many of the coagulation proteins from an inactive zymogen form to the active catalytic state. The end result here is also fibrin formation.
The description of two distinct systems for the activation of factor X is oversimplified, and there is evidence of a definite interaction between the intrinsic and extrinsic systems of coagulation. It has been demonstrated that factor VIIa of the extrinsic system can also activate factor IX of the intrinsic system, 15 and it is probable that the amount of TF generated regulates the amounts of factor X or factor IX activated by factor VIIa. 16
The combined or independent activities of the extrinsic and intrinsic pathways result in formation of factor Xa, which converts prothrombin to thrombin (IIa). In this reaction, the prothrombinase complex is formed by factors Xa, V, phospholipids, and calcium. At this point is useful to remember that platelets are actively secreting the content of α granules, which are rich in factor V and markedly increases the rate of thrombin activation. Thrombin now catalyzes the conversion of fibrinogen into fibrin, and this is accomplished in three steps. Initially, fibrin is formed in monomers and then polymers held together by weak electrostatic bonds. Finally, stabilization of the clot occurs by the action of factor XIII, which catalyzes a transamidation reaction between adjacent fibrin monomers to form the covalent bonds necessary for mechanical stability.
Now the vessel is sealed. When there is no longer a need for thrombus formation, there has to be a system that prevents the coagulation process from spreading to other sites where it is not needed. This is accomplished by an active enzymatic process occurring at the site of injury at the same time as the coagulation process.
Thrombin is the most important coagulation modulator: it activates factors V, VIII, I, and XIII and stimulates platelet recruitment. It also prevents clot spread by inducing the release of tissue plasminogen activator (t-PA) and urokinase-type plasminogen activator from endothelial cells. Thrombin also interacts with thrombomodulin and activates protein C. Generation of low levels of thrombin can, therefore, stimulate the formation of the inhibitory protein C/protein S complex and, in concert with antithrombin III (AT-III), block the complete activation of the coagulation cascade.
AT-III is the major physiological inhibitor of blood coagulation. Small amounts of thrombin or other activated coagulation factors are rapidly inactivated by AT-III, preventing undesired generalized coagulation. AT-III blocks the action of thrombin (IIa) by covalently binding to its active site. AT-III also blocks the action of other coagulation factors (XIIa, XIa, IXa, and Xa), kallikrein, and plasmin. When there is localized major injury to the vasculature, however, the amounts of thrombin and other factors generated are large enough to override the inhibitory effect of AT-III, and hemostasis takes place. AT-III potency is increased several thousandfold by heparin.
The process called fibrinolysis is responsible for degrading fibrin. The most important reaction in this process is the conversion of plasminogen into plasmin, which breaks down fibrin into small, soluble components. Plasminogen is synthesized in the liver and is present in the blood, saliva, and tears. The major source of plasminogen activator in physiological thrombolysis is the vascular endothelium. 17 In this process, t-PA, plasminogen, and plasmin bind to fibrin of the forming clot, and fibrinolysis occurs in a three-step fashion, resulting in the production of fragments D and E. The plasmin degradation products of cross-linked fibrin can be distinguished from those of fibrinogen by the presence of the D dimer, which is derived from the cross-linked gamma-gamma chains of fibrin. The final breakdown products also inhibit platelet adhesion and aggregation.
Laboratory Tests of Hemostasis
Anesthesiologists have to make decisions about coagulation tests in their patients on a daily basis. Some questions they are faced with are which tests should be ordered, when to order them, and what to do with an abnormal result, even if the test was ordered by another care provider. To clarify these questions, the most common coagulation tests used in clinical practice are briefly reviewed.
The evaluation of platelet function is still mostly restricted to quantitative rather than qualitative assays. These tests are most often performed by electronic particle counters. The accuracy and precision of these determinations are linear in most instruments for counts ranging from 103 and 106 platelets/mm3. Quantitative manual confirmation usually is not necessary, but qualitative morphological assessment is desirable, especially when certain conditions, such as pseudothrombocytopenia caused by ethyenediaminetetraacetic acid-induced clumping of platelets, can fool an electronic counter.
The clinical significance of different degrees of thrombocytopenia varies. In general, spontaneous bleeding rarely occurs with a platelet count >20,000/ mm3. A patient with a platelet count between 20,000 and 100,000 will not have spontaneous bleeding but may have bleeding associated with surgery and trauma. A platelet count >100,000/mm3, assuming that function is normal, is adequate for any challenge of hemostasis. It is important to remember that young and rapidly turning-over platelets are hemostatically very effective, so disorders causing peripheral platelet destruction with a normal bone marrow may cause less bleeding than the thrombocytopenia of inadequate bone marrow production.
The lack of any tests to assess platelet function led to the development of the platelet-activating clotting factor test (PACT) (Coulter Electronics Hialeah, FL) and the PFA100 (Dade International, Miami, FL). Several studies evaluated these devices with mixed results. Despotis and coworkers 18 reported some correlation with PACT results and postoperative bleeding in cardiac surgery. Ereth and colleagues, 19 however, found PACT to be less sensitive than thromboelastography in patients undergoing cardiopulmonary bypass.
For a long time, it was thought that this test reliably assessed the integrity of the primary hemostatic mechanisms, namely platelets and endothelium. Over the last decade, however, numerous studies clearly indicated that bleeding time is not an effective predictor of bleeding during surgery, particularly in patients with a negative bleeding history. Despite efforts to standardize the method, performance of bleeding time entails so many technical variables that it cannot be consistently performed. The test is also affected by changes in hematocrit and skin hydration, and it has also been shown that the bleeding time from an incision in the forearm does not reflect bleeding elsewhere in the body. Consequently, the bleeding time has no role in the preoperative evaluation of a patient with a negative personal and family bleeding history, 20 even in those patients taking aspirin. 21
Prothrombin time (PT), described by Quick in 1935, measures the final common pathway of coagulation as well as the presence of factor VII, therefore including the entire extrinsic pathway. The PT becomes prolonged when levels of factor V, VII, and X fall below 50% of normal and when prothrombin levels are less than 30% of normal. The PT is used to monitor warfarin therapy. The PT is also affected by gross changes of fibrinogen and very high levels of heparin. Prolongation of the PT is an unspecific indicator of an extrinsic or common pathway defect. It may be due to a deficiency of a single factor in congenital coagulopathies or a combination of factors, as is the case in most of the acquired disorders of coagulation. The clinical significance of this test is discussed later.
Activated Partial Thromboplastin Time
Activated partial thromboplastin time (PTT) measures the integrity of the entire intrinsic pathway of coagulation and thus is sensitive to deficiency of all of the factors other than VII and XIII. The activity level of the factors has to be less than 30% of normal to alter the test. This test is prolonged in the presence of certain anticoagulants and is often used to monitor heparin therapy.
This test is sometimes included in routine preoperative screening. Fibrinogen is an acute-phase reactant, and its level may be elevated above normal in response to trauma or inflammatory states. There are patients who have hereditary deficiencies of fibrinogen, but the incidence of this abnormality is extremely low.
Common Coagulation Disorders and Their Relationship With Commonly Used Tests
When the values of both the PT and PTT are normal, it is highly unlikely that there are clinically significant deficiencies in any of the coagulation factors. Principal causes for prolongation in the PT or PTT include disseminated intravascular coagulation (DIC), liver disease, warfarin therapy, heparin therapy, vitamin K deficiency, a congenital factor deficiency, factor VIII deficiency secondary to von Willebrand's disease, a lupus anticoagulant, or specific coagulation factor inhibitors. Each of these entities is now reviewed as well as the likelihood of their diagnosis by PT and PTT testing in preoperative patients with a negative bleeding history.
There are a number of well-known predisposing factors to DIC: cancer, burns, shock, sepsis, and massive transfusion. All of these conditions should be readily apparent; therefore, abnormalities in coagulation tests are unlikely to be the first manifestation of DIC. 22 The possibility of uncovering unsuspected liver disease with just PT/PTT testing is relatively low. By the time liver dysfunction is severe enough to cause coagulopathy, there are enough clinical signs and elevation of standard liver function tests such as bilirubin, alanine aminotransferase, aspartate amino transferase, and alkaline phosphatase. The likelihood of identifying unsuspected warfarin or heparin therapy is vanishingly small; it is rare that individuals surreptitiously use these medications or withhold this information from their care givers. Vitamin K deficiency usually prolongs the PT and, if very severe, also affects the PTT. If a patient has not ingested a regular diet, the diagnosis of vitamin K deficiency, by means of PT and PTT testing, is a possibility. If a history of a normal diet is elicited, however, then preoperative screening is not likely to be productive for the diagnosis of this condition.
When there is deficiency of factors VIII, IX, XI, or XII, the most common of which are deficiencies of factors XI and XII, the PTT will be prolonged and the PT normal. Factor XI deficiency is most common in individuals of Jewish descent and is not associated with clinical bleeding. Many of them never have abnormal bleeding, even when challenged with surgery, and they do not need fresh-frozen plasma (FFP) before any invasive procedure. In the end, even though the PTT can detect deficiency, the risk of abnormal bleeding depends not on the result of the test but on the presence or not of a personal or family history of bleeding. Factor XII deficiency is very common among individuals of Asian descent. As expected, factor XII deficiency prolongs the PTT. Interestingly enough, even when there is complete absence of this factor, there is not an associated increased risk of bleeding.
Factor VIII and factor IX deficiencies are responsible for hemophilia A and hemophilia B, respectively. It is extremely unusual to first diagnose these entities before surgery, because these patients are predisposed to bleeding, even with minor trauma, very early in life. Mild factor VIII deficiency may also occur in patients with von Willebrand's disease. If the bleeding history is significant for only mild episodes or easy bruisability, this deficiency may be unsuspected preoperatively. The PTT may be useful in the diagnosis of unsuspected von Willebrand's disease. Deficiencies in factor VII prolong the PT but not the PTT. Deficiencies in factor II, V, and X prolong both tests. All four of these deficiencies are so unusual that it would cost millions of dollars to identify the rare individuals with these coagulopathies during preoperative screening.
Lupus anticoagulant will sometimes prolong the PTT. Patients with this disorder or an anticardiolipin antibody have an increased risk of thrombosis, and there is not a bleeding risk, unless these antibodies cause significant thrombocytopenia or low thrombin levels and, consequently, a prolonged PT. Patients will have a prolonged PT only if the PTT is also affected. For this reason, the PTT is more valuable than the PT in identifying this condition. A PTT prolongation resulting from lupus anticoagulant may inappropriately cause a patient's surgery to be canceled or prompt treatment with FFP.
The final entity to consider in the differential diagnosis of a prolonged PT and PTT is a coagulation factor inhibitor. Inhibitors to factors VIII and IX can occur spontaneously and will prolong the PTT. At Massachusetts General Hospital, only six individuals per year are identified with spontaneous factor VIII inhibitor relative to thousands undergoing surgery during the same period. Inhibitors to other factors are several hundredfold less common.
In summary, for patients with a negative bleeding history, the principal value of the PTT is to detect factor XI and XII deficiencies, the presence of lupus anticoagulant, or factor VIII deficiency associated with von Willebrand's disease. All unsuspected causes of a prolonged PT with a normal PTT are extremely uncommon, which leads us to reconsider the value of this test in the preoperative setting. 23
Approach to the Asymptomatic Patient
The most important thing to remember about the asymptomatic patient is that no test can substitute for a detailed history and a careful physical examination. The history should include the following questions: Is there any personal or family history of a bleeding tendency? Has the patient undergone surgery or dental extractions previously? Is there any history of hematuria, gastrointestinal hemorrhage, easy bruising, hemarthrosis, metromenorrhagia, or epistaxis? Is there any history of cancer or collagen vascular disease? What medications is the patient taking or has taken recently? Is the patient on any special diet? A careful physical examination should note any adenopathy, splenomegaly, or hepatomegaly. Hepatic insufficiency should be assessed by seeking signs of jaundice, telangiectasias, gynecomastia, testicular atrophy, or any other stigma of liver disease. Assessment of the skin and mucosal surface is mandatory.
In a review of preoperative testing, Munro and colleagues 24 found that the platelet count was abnormally low in less than 1.5% of patients, and further work-up rarely revealed any significant hematological abnormality. The PT and PTT were abnormal in 4.8% and 15.6% of patients, respectively. Suchman and Mushlin 25 studied a large number of patients and found that the PTT had no ability to predict the occurrence or absence of hemorrhage in a low-risk group, but it was a predictor of modest strength in the high-risk group. Their data justify limiting preoperative coagulation screening to patients with active bleeding, known or clinically suspected bleeding disorders (including use of anticoagulants), liver disease, malabsorption, malnutrition, or other conditions associated with acquired coagulopathies, and to patients whose procedures may interfere with normal coagulation. 25
No series in the literature reports that routine preoperative coagulation screening in an asymptomatic population yields any benefit or significantly changes the anesthetic plan or outcome. When an extreme challenge to the hemostatic system is anticipated (e.g., cardiopulmonary bypass), all patients receive a platelet count, PT, and PTT preoperatively if for no other purpose than to delineate baseline values. The ordering of coagulation tests by different practitioners varies widely, and it is sometimes dependent on the comfort level of a particular surgeon or anesthetist. At Massachusetts General Hospital, a combination of common sense and clinical practice has led to the development of the algorithm shown in Figure 2. 26
Coagulopathy of Massive Transfusion
Massive transfusion is defined as the replacement of one blood volume within a 24-hour period. A blood volume roughly equates to 12 units of packed red blood cells for a 70-kg patient. Massive transfusion is associated with many complications that include, but are not limited to, hypocalcemia, hyperkalemia, hypothermia, acidosis, and coagulopathy. In this section, coagulopathy associated with massive transfusions and the unique challenges that it presents to the anesthesiologist are discussed.
Dilutional thrombocytopenia will occur predictably if enough units of blood, or any other fluid devoid of platelets, are transfused. The absolute threshold for the definition of dilutional thrombocytopenia ranges from 50,000–75,000/mm3. A classic study by Hiippala and coworkers 27 demonstrated that the platelet count will fall below 50,000 after the replacement of 2.3 times the estimated blood volume (EBV) (approximately 25–27 units of packed red blood cells in a 70-kg male). The decision to transfuse platelets in this situation should be based on the presence of clinical coagulopathy rather than a specific value of the platelet count. It is also important to realize that platelet counts may take a long time to be reported, so that the result reported may no longer reflect the current clinical situation.
Dilution of Factors
Hiippala and others 27 reported that fibrinogen was most vulnerable to dilution of the factors measured. Low levels of fibrinogen can be expected after replacement of 1.4 times the EBV. At this point, one must remember that when a patient is undergoing massive transfusion, it is because something is terribly wrong and other causes of bleeding (e.g., DIC or hypothermia) may be present besides a linear decay of factors resulting from dilution. Murray and colleagues 28 reported that clinical coagulopathy usually did not occur until more than one EBV has been replaced and the values for PT and PTT are greater than 1.5 times control. In the operating room, however, rapid access to PT and PTT results may not be available, and the decision to replace factors in the form of FFP must be dictated by the clinical situation.
Hypothermia is defined as a body temperature of less than 35°C. In the setting of massive transfusion, this is a dreaded complication that can lead to a “bloody cold cycle,” in which the transfusion of cold blood products produces a greater coagulopathy, requiring the administration of more cold blood products, perpetuating a vicious cycle. For every unit of red blood cells transfused at 4°C, the body temperature decreases by approximately 0.25°C. 29 This complication can be prevented by the use of appropriate blood warmers, some of which can deliver up to 1.5 L/min with minimal decreases in body temperature. 30,31
Practical Tips for the Management of Massive Transfusion
Sometimes the need for massive transfusion is totally unexpected and takes the team by surprise, as when a laparoscopic trocar inadvertently penetrates a major abdominal vessel. Usually, however, the need for transfusion can be anticipated by the clinical situation at hand (e.g., liver transplantation or massive trauma).
Preparing in advance goes a long way. The operating room designated for emergency cases should contain all the equipment necessary to obtain intravenous access and warm transfused fluids. Remember that it is much better, and easier, to prevent hypothermia than to restitute normothermia.
Call for help. These cases require not only a clear mind but also extra pairs of hands. Assign every member of your team a specific task, and try to look at the whole picture and the surgical field.
Set up your coagulation testing strategy so you are never far behind. It is a good idea to send baseline values and then send another set for every six units of packed red blood cells transfused.
Think of hypocalcemia (i.e., citrate toxicity) if you find yourself transfusing more than one unit of packed red blood cells approximately every 5 minutes.
Always look at the electrocardiogram. This may sound simplistic, but it is your best tool to detect life-threatening hyperkalemia rapidly.
Even though the risk for acquiring a transfusion-related infection has decreased tremendously, the cost and availability of blood products make it mandatory for us to consider seriously whether a patient is actually going to benefit from receiving blood products. A thorough evaluation of a patient's bleeding risks is indispensable. Preoperative coagulation testing should be goal oriented and cost-effective. From this discussion, it is clear that there is no such thing as a standard coagulation panel. The preoperative coagulation tests performed, if any, should be tailored to each patient based on the clinical characteristics most likely to predict perioperative bleeding.
1. Cardigan RA, Mackie IJ, Machin SJ: Hemostatic-endothelial interactions: a potential anticoagulant role of the endothelium in pulmonary circulation during cardiac surgery. J Cardiothorac Vasc Anesth 1997; 11:329–336
2. Furhgott RF, Zawadzki JV. The obligatory role of endothelial cells in the relaxation of arterial smooth muscle by acetylcholine. Nature 1980; 280:373–376
3. Ignarro LJ, Buga GM, Wood KS, et al. Endothelium-derived relaxing factor produced and released from artery and vein is nitric oxide. Proc Natl Acad Sci U S A 1987; 84:9265–9569
4. Rapoport R, Murad F. Endothelium dependent and nitrovasodilator-induced relaxation of vascular smooth muscle: role of cyclic GMP. J Cyclic Nucleotide Prot Phos Res 1983; 9:281–296
5. Lincoln TM, Cornwell TL: Intracellular cyclic receptor protein. FASEB J 1993; 7:328–338
6. Radomski MW, Palmer RM, Moncada S. The role of nitric oxide and cGMP in platelet adhesion to vascular endothelium. Biochem Biophys Res Commun. 1987; 148:1482–1489
7. Fitzgerald GA, Brash AR, Falardeau P, et al. Estimated rate of prostacyclin secretion into the circulation of normal man. J Clin Invest 1981; 68:1272–1276
8. Bolton TB, Lang RJ, Takewaki T. Mechanism of action of noradrenaline and carbachol on smooth muscle of guinea pig anterior mesenteric artery. J Physiol 1984; 351:549–572
9. Feletou M, Vanhoutte PM. Endothelial dysfunction: a novel therapeutic target: the alternative EDHF. J Mol Cell Cardiol 1999; 31:15–22
10. Monbouli J-V, Vanhoutte PM. Endothelium derived hyperpolarizing factor(s): update in the unknown. Trends Pharmacol Sci 1997; 18:252–256
11. Pearson JD, Gordon JL. Nucleotide metabolism by endothelium. Annu Rev Physiol 1985; 47:617–627
12. Warn-Cramer BJ, Maki SL, Rapapor SI. Heparin releasable and platelet pools of tissue factor pathway inhibitor in rabbits. Thromb Haemost 1993; 69:221–226
13. Siess W. Molecular mechanism of platelet activation. Physiol Rev 1989; 69:58–178
14. Dephino RA, Kaplan K. The Hermansky-Pudlak syndrome: report of three cases and review of pathophysiologic and management considerations. Medicine 1985; 64:192–202
15. Osterud B, Rapaport SI. Activation of factor IX by the reaction product of tissue factor and factor VII: an additional pathway for initiating blood coagulation. Proc Natl Acad Sci U S A 1977; 74:5260–5264
16. Zur M, Nemerson Y. Kinetics of factor IX activation via the extrinsic pathway: dependence of Km on tissue factor. J Biol Chem 1980; 255:570–577
17. Loskutoff DJ, Edgington TS. Synthesis of a fibrinolytic activator and inhibitor by endothelial cells. Proc Natl Acad Sci U S A 1977; 74:3903–3907
18. Despotis JG, Levine V, Filos KS. Evaluation of a new point of care test that measures PAF-mediated acceleration of coagulation in cardiac surgical patients. Anesthesiology 1996; 85:1311–1315
19. Ereth MH, Nutall GA, Klindworth JT, et al. Does the platelet activating clotting test predict blood loss and platelet dysfunction associated with cardiopulmonary bypass? Anesth Analg 1997; 85:259–264
20. Peterson P, Hayes T, Arkin C, et al. The preoperative bleeding time test lacks clinical benefit. Arch Surg 1998; 133:134–139
21. Ferraris VA, Swanson E. Aspirin usage and perioperative blood loss in patients undergoing unexpected operations. Surg Gynecol Obstet 1983; 156:439–442
22. Myers ER, Clarke-Pearson DL, Olt GJ, et al. Preoperative coagulation testing in a gynecologic oncology service. Obstet Gynecol 1994; 83:438–444
23. Van Cott EM, Laposata M. Turnaround times. Boston Massachusetts General Hospital, 1996; 5:1–3
24. Munro J, Booth A, Nicholl J. Routine preoperative testing: a systematic review of the evidence. Health Technol Assess 1997;1:i–iv, 1–62
25. Suchman AL. Mushlin AI. How well does the activated partial thromboplastin time predict postoperative hemorrhage? JAMA 1986; 256:750–758
26. Sady S, Sweitzer B. Hematologic issues. In: Sweitzer B, ed. Handbook of preoperative assessment and management. Philadelphia: Lippincott Williams & Wilkins, 2000:159–195
27. Hiippala ST, Myllyla GJ, Vahtera EM. Hemostatic factors and replacement of major blood loss with plasma poor red cell concentrates. Anesth Analg 1995; 81:360–365
28. Murray DJ, Olson J, Strauss R, et al. Coagulation changes during packed red cell replacement of major blood loss. Anesthesiology 1988; 69:839–845
29. Sohmer PR, Scott RI. Metabolic burden of massive transfusion. Prog Clin Biol Res 1982; 108:273–283
30. Yamauchi M, Nakayama Y, Yamakage M, et al. A preventive effect of fluid warmer system on hypothermia induced by rapid intravenous infusion. Masui 1998; 47:606–610
31. Minore WS. The use of the haemonetic rapid infusion system (RIS) has dramatically reduced the duration and frequency of significant intraoperative hypotension and hypothermia that can occur in major cases. Resuscitation 1997; 35:273–274
© 2001 Lippincott Williams & Wilkins, Inc.